
Go Starship go!
Why use hydrogen when Musk has shown methane is far superior? Honestly, just coz they had some old shuttle engines lying around...
It's looking like Starship/Super Heavy will be the first, despite the FAA's attempts to delay it.
It was second time unlucky for NASA as its Space Launch System rocket remained rooted to its Florida launch pad following a second scrub of its Moon mission. Once again, the countdown was fraught with drama when over-pressure warnings were triggered in the liquid oxygen and hydrogen lines as engineers chilled down the plumbing …
methane is certainly more dense than H2 and can be liquified more easily, but you get the hottest burn with H2+O2. Hotter means better thrust efficiency. This is why engines that only operate in space (for Apollo) use H2+O2 whereas engines that act as 1st stage boosters often use other fuels for reliability and physical size (H2 tanks need to be much larger).
[Engineers doing the math, figuring out what works best]
To double thrust you must do one of two things: double the mass flow rate, or quadruple the energy from the burn. At some point there is a maximum benefit, and I'm pretty sure rocket surgeons (heh) go for that. I would. Then you look at system and operating costs, materials, safety, yotta yotta. It's all figgered out.
And the reason that the Shuttle main engines used hydrogen/oxygen is that they were based on the Saturn V 2nd stage engines.
Engines in space need to be Hydrogen/Oxygen cos it gets most zoom/blue-whale which is important if you have to lift all the fuel into space.
But you get less total zoom than oxygen/aga-fuel which is why the Shuttle needed the big fireworks, which fortunately caused no problems, and is the reason that everyone else uses oxygen/aga 1st stages (including the Saturn V)
Lots of rockets not using LH use boosters as well to take off. There were alternatives Shuttle designs that used a first stage and didn't use boosters.
I don't remember many issues filling the Shuttle main tank - one would think NASA knows what it is doing - in the Saturn V itself two whole stages had to be filled with LH for launch anyway - so if you get issue loading tanks it doesn't really matter where they are.
One reason NASA get on the Moon and far beyond Mars was because it mastered LH technology (Centaur included). Maybe today it could be replaced by something less - but we have to see yet one rocket achieving it.
"I don't remember many issues filling the Shuttle main tank - one would think NASA knows what it is doing - in the Saturn V itself two whole stages had to be filled with LH for launch anyway - so if you get issue loading tanks it doesn't really matter where they are."
Shuttle averaged a scrub a launch I think... And the H2 lines were always difficult - H2 is just hard to handle.
Er, the Saturn V used highly refined fuel oil + liquid O2 for the first stage. Basically high quality diesel/jet fuel.
If you limit yourself to fuels that are relatively easy/safe to handle/allow throttling/ then the hydrocarbons give you the most energy/volume. Very important when air resistance is a factor. It is only when one is out of the atmosphere that the energy/weight advantage of hydrogen comes into play.
The Mickey Mouse kludge of the space shuttle design was a political decision, not an engineering one. It has been argued that the primary purpose was to require the maximum manpower to operate in order to keep the knowledge base employed in the interim between Apollo and the proposed manned Mars missions.
well space shuttle focused on re-usability and was a bit of an experiment in that regard.
SO yeah SRBs plus "one big tank" instead of multiple stages, and yeah the shuttle engines had to work well in a vacuum as well as on the ground. A compromise. Most of the aerodynamic losses were probably nn the fuel tank, though, before SRB separation. So the SRBs would have (probably) made up for any advantages of using fuel oil at those altitudes.
I think the same logic applies to Artemis, use the SRBs to assist where H2+O2 makes less sense, but get max performance out of the whole system trying to use just one stage + boosters to achieve orbit.
more of a philosophical decision, perhaps?
Wrong.
The Space Shuttle Main Engines run a Fuel Rich Staged combustion cycle. The J-2s ran a straight gas generator. Totally different.
You might like to think of the Saturn V US as Merlins, and the SSME's as Raptors, although the SSME has substantially better Isp and poorer T/W ratio. LH2/LO2 is simply the highest Isp combo that's (relatively) safe to handle.
And of course the SSME's each made orbit at least 35 times, whereas Starship has yet to reach orbit, although I'm sure we're all expecting it to make orbit at least once before the end of the year.
> this scrub was on the lines pumping the fuel IN to the rocket.
The worrying thing is that this (loading H2) is something that has happened many times before on this hardware. NASA should not be discovering that their equipment is broken at this stage in the preparations.
And the one part that cannot be tested is the SRBs.
To be entirely factually correct the actual scrub was due to what they have now deduced to most likely be a sensor showing an incorrect engine temp. They hope. Fingers crossed. But they can't actually change the temp sensor or verify that it is indeed showing incorrect. So there is a change it IS showing correct, there is a problem with the H2 chill valve or line leading to low flow and insufficient cooling and the engine goes kaboom in a spectacular fashion when they actually fire it.
The parts giving trouble may well be 40+ years old. The H2 tank has been there since STS 1.
One of the more interesting inventions used to monitor leaks on those systems are Chemochromatic tapes for leak detection.
Very simple and reliable detection system; though it can’t escape anyone that the ability to contain hydrogen is very difficult to achieve with NASA budget. I’m not the only one with concerns for using hydrogen elsewhere; even in blended forms.
The main problem with hydrogen is it requires an enormous tank, which gets heavy.
The core up to the bottom of the nose cone of the boosters is the hydrogen tank. That huge tank only holds 317,000lbs of hydrogen.
The tiny rest of the tank from there to the part that necks down is the oxygen tank, which is 1,860,000lbs of oxygen.
If you were flying with methane or RP-1, that tank would be a LOT smaller and more importantly, lighter. That's why the Saturn V used RP-1 in the first stage.
And no, the Saturn V didn't use hydrogen in the first stage, for that reason.
There is a lot on this in "Ignition!: An informal history of liquid rocket propellants" where the multiple trade-offs in terms of specific impulse, storage size, and handling difficulties is covered.
TL;DR H2 & O2 has the highest specific impulse of and sane-to-handle fuel choices, but it is pain to handle.
Exactly. Did I mention Musk will use methane?
Oh, and he seems to get stuff to orbit every week currently using effectively aviation jet fuel. Two billion dollars a launch (well, possible launch). Two billion dollars -- seriously?
Maybe Musk has thought about this and decided hydrogen is pants. What's the word I'm looking for? Oh yeah, I remember, doh!
It didn't take the Tweetmeister to decide that H2/O2 first stage is pants, everyone with a copy of Rocketry for Dummies knows that and prefers methane, aga-fuel, or if you're feeling brave, UDMH or if you're suicidal, chlorine trifluoride+just about anything
H2/O2 is what you use if you're told to make a rocket with no time, no budget but access to the NASA parts bin
Yea funny how starting a massive project by cutting corners makes it take longer and cost more on the long run.
Anyway we can now start HS2 cos we found some spare Brunel gauge engines in a museum so by designing around reusing them we assume the cost will be halved
Yes, but the main advantage of methane is not bigger bang for your mass, but a lower volume and difficulties of handling it (hydrogen gets in everywhere, making metals brittle) and more importantly if you are designing a new engine you can make it full-flow staged combustion so turbo pumps have an easier and cheaper job.
Beryllium and oxygen? The melting point of Be (~1300C) and the boiling point of beryllia (~4000) may be a bit of an engineering problem if you want to pump fuel, and have the exhaust leave as a gas (melting point of liquid ~2500C, which would still be tricky). Perhaps fluorine as the oxidiser, but then maybe lithium as the fuel? Both fluoride salts are liquid at <600C. Oh sorry, you said "sane" to handle, that takes fluorine out...
To be nearly serious: High impulse/low volume systems like those based on hydrazines/nitrogen oxides are useful in (extra)orbital craft, but their cost, handling difficulty, and toxicity make them impractical for launch vehicles. As an aside, I worked on this with some of them, including 95% H2O2...
One of the reasons for using methane is that the tanks end up about roughly equal size. mass and temperature whilst still having a decent ISP and total thrust.
The temperature part matters more than you might think. Significant problems have been caused by tank/plumbing interactions between LOX/RP-1 and LOX/LH2 at various points in the past
Obtrivia: there are significantly more hydrogen atoms in a gallon of RP-1 (or any other liquid hydrocarbon fuel at room temperature) than a gallon of liquid hydrogen
Quick and Easy for whom?
I'm minded of a bunch of frankenstein projects I had to deal with where blowhard techs would exclaim "It's all proven stuff", then tie things al together using a rat's nest of other shit which wasn't tested properly and interfered with all the existing equipment, vs a simple redesign of the existing stuff to rescale it slightly ("Too hard, I forbid you to work on it")
Unfortunately such people end up in manglement and sales
Yep, nobody actually cares about the moon. SLS is going to the moon because putting persons on the moon is the simplest achievable goal that the public understands.
Cos wasn't JFK cool and weren't the 60s awesome (if you were a square-jawed white guy fighter pilot)
So we are going to do that again, only this time one of the persons will be an ethnic and one will be a women (although all will be squared-jawed ex-fighter pilots)
Mars has CO2 so we can make methane locally, there is no real advantage in landing on the moon and taking off again to get to Mars
There's a HUGE reason to go back to the Moon first, and that reason is...
Oops.
If something catastrophic goes ping on the Moon, everyone can evac back to Earth in a lifeboat craft in a couple of days. If you're on Mars, it may take a year or the better part of a decade depending on where the two planets are in orbit. Build on the Moon, figure out how to do it all while relatively close to Earth, figure out your failure points, THEN go to Mars. And who knows, while we're working the problems out perhaps we'll make a quantum leap in transportation technology and Mars becomes an hour-long trip even at the furthest point in orbit ftom Earth.
"If something catastrophic goes ping on the Moon"
Then the odds are pretty good that you're stuck there using whatever you have onhand
The lifeboat theory mostly falls apart when you look more closely at it. If you retain the resources to use it, then you don't need it
And who knows, while we're working the problems out perhaps we'll make a quantum leap in transportation technology and Mars becomes an hour-long trip even at the furthest point in orbit ftom Earth.
You know that a quantum is the smallest possible value of something, right?
That said, the furthest distance from Earth to Mars is 400 x 10^9 m, so your one hour trip would be at an average of 37% of the speed of light. Good luck with that,
Moon has a number of downsides including the abrasive dust issue.
It's been argued for a long time that if we can't go to Mars direct then we should target mining asteroids as this would be _easier_ (and far less dangerous) than the Moon (in fact the argument is that Asteroids should be first, planets later)
Ceres is one big iceball containing more water than the entire Earth's surface (oceans, rivers and atmosphere), others appear to have significant quantities of iron and other metals (virtually all metals mined at the earth's surface were delivered by asteroid strikes)
The Moon has plenty of oxygen bound to light metals, a limited amount of hydrogen mostly in sheltered locations near the poles and a shortage of carbon. Using lunar oxygen to get off the Moon is a massive winner once you set it up. Using Lunar oxygen to get from Lunar orbit back to the Moon is still a clear winner compared to oxygen from Earth. Lunar oxygen for a trip anywhere else does not compete with oxygen from Earth at Starship prices although it would be competitive against Earth oxygen delivered by SLS - if anyone had that much money and there actually was an unassigned SLS available for launch.
The simple answer for the fuel to leave the Moon is to bring it from Earth with Starship. A more complicated solution would be to bring carbon from Earth and combine it with hydrogen from the Moon to get methane for a Starship. A simplistic way to look at cost benefit is hydrogen has a relative mass of 1 and gets you 1 bond. Carbon has a relative mass of 12 and gets you 4 bonds. Oxygen has a relative mass of 16 and gets you minus two bonds. You need to get the total number of bonds to 0 so you need oxygen. It costs 8 per bond so it is expensive to transport. As it is abundant that makes it a good local resource. Hydrogen's mass of 1 per bond makes it cheap to transport and the difficulty of extraction makes it a difficult local resource. If you want Carbon you have to bring it from Earth and the cost of 3 per bond makes it intermediate in value between hydrogen and oxygen for transport.
Hydrogen+Oxygen have an excellent _theoretical_ effective exhaust velocity. At first sight this looks like the mass of hydrogen needed to change velocity is lower than for any other chemical rocket fuel. On closer inspection, the low density requires an enormous tank. The very low boiling point makes storage tricky. Leaky connectors will spoil your day. It dissolves in popular structural metals making them brittle and leaking through them. The low density means a huge pump is needed to get it up to pressure. The huge pump needs a powerful turbine to spin it. RS-25 engines get around this by sacrificing the theoretical high exhaust velocity by sending too little hydrogen to the combustion chamber. (Chemistry says 2H:1O. Rocket science says some extra H gives a higher exhaust velocity because the exhaust contains lighter molecules.)
All of hydrogen's problems disappear if you switch to methane. Easier handling, smaller tank, smaller turbo pump. Best mixture ratio is practical. Boiling point similar to oxygen so one propellant does not freeze the other solid. The maximum theoretical exhaust velocity is lower but it is easier to get close to the maximum and overall performance including tank and engine mass work out similar to hydrogen.
There are other choices: if you have oxygen, chlorine, aluminium, hydrogen and carbon you can make solid rockets. The Moon has plenty aluminium but not chlorine. Spin launch from the Moon has fewer technical difficulties than spin launch from Earth. A space elevator on the Moon can be built with practical materials rather than extremely long nanotubes that we cannot make on Earth. Each of these requires a big step up in Lunar traffic over the last to bring a return on investment.
If we actually build Moon Base Alpha, oxygen propellant makes sense. Methane looks interesting for Starships but the economics may not work out. Hydrogen for a Blue Moon lander would be a possibility if Blue Origin could launch to orbit but the way things are going, Blue has a long way to go to compete on price with SpaceX. At this time, it is not clear that there will be enough traffic to justify Lunar oxygen propellant but if I have massively underestimated the tourist market then I would put a small bet on a Lunar Space Elevator for my great grand children over Lunar fuels.
The difference is that SpaceX does the majority of their testing on non-flight hardware. Things like disconnect tests were performed on what basically was just a big dumb steel tank. No engines, no flight controllers, no hydraulics, bare minimum electronics for the required sensors and the few valves that were there.
They they start building from there and adding complexity as they go on top of proven hardware. This means SpaceX can afford to do multiple fuelling attempts, scrubs, RUDs etc, since it's not costing them tens of millions to billions every time they do it.
There is also huge pressure to scrub if there is any possibility of a RUD. Try looking at the options after a RUD after take off: repeat Artemis 1 in 2024 or maintain the schedule by putting a crew on Artemis 2. Options get worse for a RUD that trashes the mobile launch platform. Artemis 2 cannot launch without a replacement MLP and is incompatible with MLP being designed for Artemis 3 in 2025. Either way Starship will be making progress during that time and it is remotely possible New Glenn reaches orbit before 2025.
The main tank is rated for 22 cryo-cycles. Because of testing it has had somewhere between 10 and 20 already. There will be no more test cycles. Every 'test' will be done while the rocket is ready to launch. Imagine if the only time propellant loading worked was during a test that was not during a launch window...
The boosters are certified for 12 months from stacking (2021-01-07). A few extra days is not a problem but each extra month is a risk that inspections will show degradation. If the boosters do not get re-certified then the launch abort system, Orion capsule, service module, second stage and first stage all have to be de-stacked so the SRBs can be replaced.
It may actually be worth risking a launch despite warning signs if the main tank or SRBs are very close to their expiration date anyway. Some redundant parts of the capsule have already died of old age. The crawler transporter is struggling with the extra weight too. If it gets stuck near pad 39b it could delay Crew Cragon launches that can only depart from 39a.
because Starship wont be rated for crew launch by then, its got no in flight abort system, would you sit in that thing ? that so far has managed just the 1 successful landing from 10km without blowing up. theres a world of difference between playing around with explodable steel cans in Texas, and putting people on top of it to launch into space.
I was working at the Cambridge Electron Accelerator in 1963. Right next to the beamline I was on there was this gigantic block of iron (or steel - I didn't check its pedigree). It was a liquid hydrogen bubble chamber about ten feet tall, or three meters. It was still in its youth, and had not yet been filled with hydrogen. I graduated (1963) and went back to Minnesota. The first time they filled the chamber, very carefully, the diabolic machine exploded. That was on July 5, 1965. Obviously, the hydrogen had not behaved itself.
http://tech.mit.edu/archives/VOL_085/TECH_V085_S0235_P011.pdf
You have to be insanely sure of yourself to come anywhere near liquid hydrogen.
You have to be insanely sure of yourself to come anywhere near liquid hydrogen.
Having for some years earned my living in cryogenics, I concur. Hydrogen is by far the nastiest of all the cryogenic liquids and anyone with sense tries to avoid working with it if they possibly can. At one time it was used as a cooling stage between LN2 and LHe, but He liquefaction became easier and rebuilding exploded laboratories was becoming tediously expensive, so it was dropped. When I was in the field the only place in the UK still equipped for LH2 work was the Rutherford Lab ... remote control only, behind blast walls. And even they hardly ever used it.
I don't believe it!
The bloody thing should have taken off today.
As you already know, I'd arranged for a drinks and nibbles party at the Meldrew residence and we had to call the whole match off for the 1st launch.
The "Extra Special" sausage rolls bought by Margaret and Mrs. Warbouys expired tonight at midnight, it's now 00:40 BST.
Now NASA better start praying, we shall be suing for the cost of the rolls and disappointment caused, we estimate a law suit of at least 22 USD.
Seriously though cannot wait to see this launch. (Timing to be arranged).
PS. Bought some frozen Bahji mixed nibbles from Mr. Iceland (10% off for OAP's) will store safely without Hydrogen leaks for 12 months.
Calling off the launch when problems are noticed is always the right thing to do.
At the end of the day ever launch has the potential to become a ... (see title), and no one wants that (well maybe some Luddites and possibly few flat-earthers might, but any reasonably sane person does not).
I hope that the third attempt goes 100% A-OK, but also that the countdown is halted again - if it needs to be.
An SLS RUD on the launch pad is a fairly popular outcome among well informed human spaceflight enthusiasts. SLS is extremely expensive by design and the maximum launch rate is only once per year. It is based on space shuttle tech that averaged 1.9 scrubs per launch with a maximum of 6 scrubs. SLS scrub rate is likely to be higher because it flies less often. Reduced serviceability on the pad means increased chance that a rollback to the VAB is required - incurring a month or two of delay each time.
A determined bipartisan effort has preserved Space Shuttle contractors for a decade despite the heavy cost to human spaceflight. It is believed that a RUD would end the waste and delays. I am not convinced that a RUD alone will do the job. I think it will take the combination of Starship achieving some of its goals and an SLS RUD. Some idea of the relative costs reaching voters would also help.
Last I heard, NASA had not ruled out a launch on Tuesday (unlikely) which could only be achieved by repairing the quick disconnect arm on the pad (possibly). Returning to the VAB is sufficiently slow and difficult that there is speculation about leaving SLS on the pad for a month to reduce the chance that another two month delay causes the SRBs to die of old age. Not returning to the VAB means the flight termination system batteries cannot be replaced and they could dissolve before they are needed. This rocket was designed for cost plus activity not operational cadence. Perhaps one day voters will be a tenth as furious about this as human spaceflight enthusiasts.
No sane person? Why would a RUD necessarily be a bad outcome, given the state of the program?
It’s a test, with Pass/Fail criteria. Tests tell you the status, whether good or bad, and without hiding behind Lies You Tell Yourself. One RUD would likely not lead to program cancellation, but tell them more than anything else. A re-run in a years time…might also RUD. This is only a bad thing if you think program cancellation is a bad thing *per se*. But that’s a contractors cost-plus viewpoint.We *should* have the viewpoint that human space flight is a good goal, and whichever is the most viable program to achieve that, wins.
If you have a program that might RUD, and the earliest you can try again is a year later, and that might or might not RUD……you do not have a credible means of human spaceflight. Even if I handed you a program on a plate that could *guarantee* a successful safe launch….by giving it three or four goes per year, and just call a halt if it looks unsafe, so we only get maybe one launch per two years at a slightly random time….what use is that? What are you going to do with that “capability”? We should also really be asking - what is wrong with your program methodology that a single failed test results in a year delay? It doesn’t in other *large-scale safety-critical engineering projects*. And it doesn’t for SpaceX.
If it launches, and RUDs, it’s a very public reality check. If RUDs twice more, cancel it, and give the money to someone who can develop a real capability. Look, SpaceX developed their entire program on “test and learn”, not because they have money to burn, but because they *don’t*. If these guys spend $5bn on development without launch trying to do everything in design/simulation, and *still* have launch failures, their methodology is fatally flawed. Stop doing stupid.
So. Who would of thought it would be a NO go?
For sure I did! How many thousands went to the Booze fest for the launch? Probably too many. Great for the economy in that area. Right? NASA is mostly a costly money maker that has overspent on almost all its trials. This is the costliest so far. Still they will get even more billions. They should just scrub this one and build a new rocket with New everything with fewer bosses. Re-using parts from the Shuttle series is and was stupid to start. Retire those old engineers and get new techs in who know the new stuff. Way more reliable than old politicians of past failed projects. The old don't learn. They keep trying old tech. I knew it wouldn't work and so did Eleon Musk.
Hydrogen can and does leak through the interstices of well made, good quality metalwork. Given the construction techniques employed for building the tank were about as good as you can possibly get; avoiding welding, etc. it is presumably the joints or flanges where the problems lie.
No amount of O-ring voodoo will stop that entirely. The only decision is can you stop "enough" for it not to interfere with the operation of the system.
So, as I noted above, even with NASAs budget putting hydrogen into things is inevitably going to involve leaks. Managing it even within their closed system is hard to do.
Can you imagine what throwing hydrogen into a domestic pipe network not nearly so well maintained will do?