Thursday, March 14th, 2024 saw SpaceX attempt the third Integrated Flight Test (IFT-3) of its massive Starship / Super Heavy launch system after the Federal Aviation Administration (FAA) granted a limited launch license to the company on March 13th.
Despite SpaceX and its followers hailing the first two launch attempts as “successes”, the short-order loss of both vehicles within 4 minutes of the launch of IFT-1 and which either vehicle achieving its core milestones in IFT-2, meant that both of those flights were extremely limited in their “success”. As a result of both, SpaceX spent considerable time reviewing the launch profile for the vehicles and making changes and improvement to both the Starship craft and Super Heavy. These resulted in IFT-3 being a broadly successful – although the loss of both vehicles at different points in the flight meant it was not an unqualified success.
Following lift-off at 13:25 UTC, with an initially perfect firing of all 33 Raptor engines on the booster, the stack of rocket and starship passed through Max-Q, the period where both experience maximum mechanical stresses as they ascend through the atmosphere, within the first minute of flight.
Even so, at 2:42 into the flight, the engines on the booster shut down and two seconds later, the starship upper stage ignited all six of its engines in a “hot staging” manoeuvre, separating from the booster after the engines had fired. This went a lot smoother than evidenced in the second launch attempt in November 2023, and the booster was this time able to change direction and execute an successful “boost back” burn – using the motors to kill its ascent velocity and push it back towards the launch site.
However, it was during the boost-back that possible hints of engine issues appeared: several of those recording and reporting on the launch noted that some of the engine exhaust plumes were tinged green, indicative of one or more engines consuming itself (green indicates the copper used in the engines is being consumed), a long-term issue with the Raptor 2. Nevertheless, the booster successfully re-oriented itself and started a planned engine-first descent towards the Gulf of Mexico and a splashdown.
For this to happen, the booster needed to slow itself by a further re-lighting several engines in a braking manoeuvre roughly a kilometres above the water. Whilst three engines did ignite, two immediately failed, and the vehicle was destroyed less than 500 metres above the Gulf – although it is not clear if the flight termination system was triggered or the booster blew itself apart. At the time of destruction, it was travelling with sufficient velocity to hit the water at 1,112 km/h.
Starship went on to achieve orbit, on course for a splashdown in the Indian Ocean. Travelling at around 240 km above the Earth, the vehicle carried out a test of the “Pez dispenser” payload bay door – a slot in the vehicle’s hull at the base of the payload bay and specifically designed to eject Starlink satellites (these being almost the only payload for Starship at present). Also tested was a so-called “propellant transfer” test, shunting a small amount of liquid oxygen between the vehicles’ main and header tanks.
However, SpaceX cancelled the vehicle’s planned de-orbit burn with one of its Raptor engines and instead allowed the vehicle to “go long”, continuing along its orbital track until gravity until drag caused it to re-enter the denser part of the atmosphere for a hoped-for splashdown. In the event, and following an initially very successful re-entry, the vehicle broke apart at an altitude of around 65 km.
The orbital flight segment of the test was impressive whilst also raising questions as to Starship’s future orbital flight dynamics. Notably, throughout its half orbit of the Earth, the Starship was in a state of continuous “bbq roll”, that is, spinning around its longitudinal axis (and making it seem like the Earth was constantly looping around it on videos). Such rolls are not uncommon on space vehicles when in sunlight, as they help spread the thermal load of the Sun’s heat over the vehicle’s outer skin, preventing uneven heating (or overheating).
In this respect, Starship is especially vulnerable to such thermal stresses: it is completely reliant on cryogenic propellants which tend to revert to a gaseous state (and require venting to prevent tanks being over-stressed), and it is made of stainless steel, and extremely poor thermal insulator. This is compounded by the fact that the hull of the vehicle is also the the outer surface of the propellant tanks, so outside of the thermal protection system (TPS) tiles coating one side of the vehicle and designed to protect it during re-entry in to Earth’s atmosphere, there is next to no thermal insultation between the vehicle’s propellant reserved and the Sun, thus leaving rolling the vehicle as the simplest means of regulating internal temperatures.
Even so, the rate of roll, combined with its continuous does raise questions: was the rolling seen on this flight simply an overly precautious desire to limit thermal blooming inside the vehicle, or will it be part of starship SOP in the future. If the latter, then there are going to be some significant issues to address (how are to starships supposed to pump propellants being them in they have to roll like this once mated and the fuel to be transferred from one to the other is being exposed to a severe Coriolis effect as a result of the spin? Was the spin in this instance the cause of the planned de-orbit burn being cancelled because a smooth flow of propellants to the motor to be fired could not be guaranteed?
That said, the vehicle did perform its own mini “propellant transfer”, pumping a small amount of liquid oxygen between its own tanks. However, the overall value of this test is perhaps not as significant as some SpaceX fans have stated, given it is a long way short of the 100+ tonnes of propellants at a time that will need to be transferred between vehicles when it comes to sending the proposed Starship lunar lander to the Moon .
But leaving such thoughts aside, the one undoubted spectacular element in the flight were the initial phases of re-entry into the denser atmosphere, when cameras mounted on the vehicle’s control surfaces were able to video the build-up of super-heated plasma around the craft as it slammed into the atmosphere. While this has been filmed from within various space vehicles (Apollo, shuttle, etc.), this is the first time (I believe) it has ever been recorded from outside the vehicle going through re-entry.
Another unique element of the vehicle demonstrated prior to re-entry was the use of vented gas as a means of controlling the vehicle’s orientation. As noted above, cryogenic fuels tend to “boil off” and turn gaseous unless kept perfectly chilled. This gas must then be vented in order to prevent it becoming too voluminous and rupturing its containment tank (hence why rockets using cryogenic fuels are constantly venting gasses prior to launch following propellant loading & then having to be constantly “topped off”). However, rather than just letting go of this gas in space as they do on the ground, SpaceX channel it through a series of “cold thrusters” around the starship vehicle, enabling them to use the vented gas to “steer” the vehicle, avoiding the need for more traditional (and mass-using) thrusters systems requiring their own tanks of hypergolic propellants or gas.
While overall successful, the loss of both vehicles does mean a mishap investigation overseen by the FAA has been triggered, which may delay the planned launch of another test flight originally targeted for just a few weeks time. Even so, SpaceX are to be congratulated with the results overall, carrying the company as they do a modest step forward in the system’s development.
Hiding in Plain Sight: The Ancient Volcano on Mars
Mars is know for its massive volcanoes – notably Olympus Mons, the “king of kings”, with a base area largest enough to cover France from its Bay of Biscay coast to its border with Switzerland, and stretch from Paris to Montpellier in the south, and which it 2.5 times the height of Mt. Everest when measured from sea level, and twice the height of Mauna Kea when measured from the sea floor.
Then there are the three mighty Tharsis Ridge volcanoes straddling the Martian equator: Ascraeus Mons, Pavonis Mons and Arsia Mons. Now a team of scientists have discovered a “new” ancient volcano to rival these four in terms of its area, if not its height – and it has been hidden in plain sight throughout all of our observations of Mars.
“Noctis Volcano”, as it is currently being called, is located between the massive 4,000 km long valley system of the Valles Marineris (the so-called “Grand Canyon” of Mars, despite the fact it could swallow the Grand Canyon in terms of length, width and depth, and barely notice it had done so), and the three Tharsis volcanoes. It sits with a region called the Noctis Labyrinthus (Latin: for ‘the labyrinth of night’), a chaotic area full of valleys, graben and other formations, many indicative of the action of water at some point in the planet’s past.
Because of its relationship with the Valles Marineris and its own unique form, Noctis Labyrinthus has long been of fascination to planetary scientists and frequently the subject of observation and study, its disorderly, intersecting valleys and plateaus the subject of many photographs and images of the surface of Mars. Yet the fact it was actually hiding an ancient volcano had completely escaped notice until now – and even then the discovery was down to chance.
It was made by a team of scientists led by Dr. Pascal Lee (a former acquaintance of mine in fact, who once took us on a tour of the facilities at NASA’s Ames Research Centre, where he is a part of the Mars Institute). They didn’t actually realise they’d stumbled on the volcano at first, as they were following-up on previous studies they’d made of the area, as Pascal explains:
We were examining the geology of an area where we had found the remains of a glacier last year when we realized we were inside a huge and deeply eroded volcano.
– Dr. Pascal Lee, SETI Institute Planetary Scientist
Reviewing their data and using data gathered by NASA’s Mars Global Surveyor (MGS) mission, the agency’s Mars Reconnaissance Orbiter (MRO) and Viking missions, together with data gathered by the European Space Agency’s (ESA) Mars Express (MEX) orbiter, the team realised the central area of Noctis Labyrinthus represented an ancient volcano roughly 9 km in elevation and some 450 km across – just 150 km smaller across than Olympus Mons.
The volcano’s enormous size and complex modification history indicate that it was active for a very long time. Furthermore, in its south-eastern part lies a thin, recent volcanic deposit beneath which glacier ice is likely still present. These combine to form a very powerful case for the area to potentially have had the right conditions for basic life to be kick-started: minerals, chemicals, heat, and water. Further, although the exact age of the volcano has yet to be determined, the fact that it is so old and so subject to subsequent geological activity means that it could also offer significant insights to the geological history of Mars regardless as to whether it was once a cradle for microbial life.
It’s really a combination of things that makes the Noctis volcano site exceptionally exciting. It’s an ancient and long-lived volcano so deeply eroded that you could hike, drive, or fly through it to examine, sample, and date different parts of its interior to study Mars’ evolution through time.
– Dr. Pascal Lee, SETI Institute Planetary Scientist
One of the reasons this region has long fascinated scientists – outside of the presence of the remarkable Tharsis volcanoes, the uniqueness of the terrain and the magnitude of the Valles Marineris – is the fact that many of the deep canyons and valleys with the region are deep enough to have somewhat denser atmospheric pressure than the rest of the planet and can exhibit their own weather patterns – clouds, frost, etc.
In this, “Noctis Volcano” is further interesting as it is close to the Martian equator, thus receiving plenty of sunlight all year round to help warm it. In addition the glacial deposits are close enough to the region’s surface that they might be reasonably be reached, providing not just the opportunity to study samples, but to actively use the ice to generate air, water and fuel for use by human explorers.
Voyager 1 Update: Clue to a Cure
At the end of 2023, I covered the issues currently being encountered with Voyager 1, our first interstellar space probe (see: Space Sunday: 1,000 sols and counting), noting that while the craft can receive communications from Earth and act on them, its own transmissions are garbled. In a more recent update, I noted that NASA had confirmed they’d traced the issue to the flight data subsystem (FDS), the computer responsible for packaging science and engineering data before it is sent to Earth by the spacecraft’s telemetry modulation unit (TMU).
However, trying to rectify the issue was proving difficult for a number of reasons including Voyager 1’s age (which obviously includes the age of its computer systems in terms of having all the specialised knowledge available to plumb their 47-year-old depths), and the sheer distance between the vehicle and Earth, which requires nigh-on 48 hours for a single 2-way communication.
At the start of March however, a NASA engineer not directly involved with the mission, but working on NASA’s Deep Space Network (DSN) which handles all communications between NASA centres on Earth and their robotic missions, realised that outlier data in the gibberish stream of binary data Voyager 1 sent in a packet of data received on March 3rd, 2024 was not actually gibberish at all.
The data has been sent in response to a “poke” mission control had sent the spacecraft a few days previously, designed to see of the FDS would send different gibberish via the TMU. After the transmission had been decoded (which itself took almost a week), the DSN engineer asked if he could eyeball it and realised the outlier data was actually a complete readout of the FDS’ entire memory, including performance instructions and code values that can change either if the spacecraft’s status changes unexpectedly or if it is commanded to change its status – it was just in the wrong data stack.
With this discovery, the hope is that the data can now be compared with similar memory dumps sent by the FDS prior to it starting to send gibberish. If discrepancies can be found between the two sets of information, they might either directly point engineers towards the underpinning cause of the problem with the FDS, or at least help them narrow down the hunt for the cause. And if this proves to be the case, it might be possible to determine what, if anything, can be done to get the FDS talking common sense again and allow Voyager 1 to resume its science mission.
Inara Pey: Living in a Modemworld 
I have to say, while your detailed reporting regarding the SpaceX launches make for interesting reading (it’s always great to get answers to the many “whys” regarding the choices made by SpaceX on their spacecraft, and understand why some choices, while cheaper, may have serious consequences & impact on how the mission fares), it’s those little tidbits about the communication with the Voyager that simply fascinate me!
I suppose it’s because I have a computer-related background, and what these guys at NASA are doing is simply mind-blowing — remotely debugging software on a still-working computer from half a century ago (!!!) at the slowest round-trip times ever experienced. Nobody, at the time the spacecraft was launched, could even remotely imagine that this would be possible at all.
Kudos to NASA who knows how to protect their on-board computers so well that they resist after so many decades (decades!) of being exposed to the cold, dark void of space, where probably not even a lot of cosmic radiation from the Sun reach the vehicle. But our communications reach it, and it answers back, “garbage” or “no garbage”. Of course, I’m aware that the software was remotely upgraded several times (even that is fascinating in itself), bugs were fixed, etc. and this was all part of the spacecraft’s design, but I’m really sure that nobody, back then, could imagine that this could go on for so long — and so far away.
I mean, I remember discussing the Voyagers at high school, and back then, we were already being told that the probes had technically over-reached their limits and their extraordinarily successful missions were essentially completed years before, and now there is nothing much more to discover, and/or deal with, since the craft would be essentially a “dead” body of no practical use except to eventually serve as a memorial to “this is what Earth technology could accomplish in the 1970s”. Not exactly a footnote in history — most of the planetary imagery that illustrated our physics and geography books came, after all, from the then-recent images from the Voyagers — but something, like the garbage that the Apollo mission left on the Moon, that we could essentially “forget” that it was still around.
Instead, the NASA engineers and programmers managed to keep the spacecraft going on with a certain degree of communication, from telemetry data from some scientific data from equipment still working after half a century, and have done all the possible tricks to keep, in a sense, the mission “alive”. It did explore the limits of the solar system and cross over to what we technically now consider “interstellar space” — and while it was expected that this would happen “sooner or later” (unless the craft blew up or something like that, such as unexpectedly crashing on a previously unknown, very dark planetoid.
Oh well. I have no idea if the Voyager is able to continue much further, sending back “status reports” as it has been doing for what now we can say “it’s been around for two human generations, who can predict for how long this will be able to continue”…
LikeLike
I’m afraid SpaceX starship is liable to be cropping up a lot in Space Sunday, as I’m working on covering the system from an objective perspective: will it work (probably); can it become a worthwhile payload launch system (possibly); is it a viable lunar landing system (no, not really); will it be capable of carrying people to Mars (highly unlikely, and certainly not in the 100-people-per-ship nonsense Elon Musk claims). Drat… I think I just did the entire series right there! 🙂
The Voyager missions (and the Pioneer missions with preceded them and which are also still going) really are remarkable. Hopefully, while blind and running on minimal power, Voyager 1 can be coaxed back to talking common sense – but full kudos to the DSN engineer who spotted the valid data among the nonsense!
LikeLiked by 1 person
Time for Elon Musk to step in to the next multi-billionaire project: the Space Elevator! 😉
That would not only solve all his current issues — launching more and more satellites for Starlink, and keep replacing those that get lost/damaged or reach end-of-life — but also place very easily, and cheaply, as many people in orbit as he wants.
Then all he needs is a very simple spacecraft, point it somewhere in the direction of Mars, and give it a kick 🙂
Of course, on the Martian side, his army of robots must already have built another space elevator as well. The good news is that once you have the first space elevator on Earth, all the others are much cheaper to build (and the one on Mars will also only have to deal with a third of the gravity and way, way less air friction on the lower parts).
Rockets, IMHO, are a “jack-of-all-trades-but-master-of-none” technology for spacecraft: they can do every possible mission (at least in the solar system), but they’re extremely expensive, extremely inefficient, and cause extreme damage to the environment. Rather, to do real space exploration & colonisation, you’ll need a plethora of different technologies, working together.
To launch cargo into orbit, unless we develop anti-gravity, the space elevator is our best bet. It can be so efficient (perhaps not on the first generation, but…) that the only thing the motor powering the “conveyor belt” needs to do is to overcome air friction close to ground — possibly you can keep it running from solar panels on the anchor point. Building it (and, of course, maintaining it) is the only real cost, and that’s paid only once. It’s like building a bridge: very complex to design, highly expensive to build, taking a long time until completion, but… once it’s done, it’s done, and it can carry all required traffic essentially for all eternity (minus, of course, some maintenance costs and minor repairs here and there) across two riverbanks, without any extra cost (unlike a ferry boat).
Once in orbit, yay — no more air friction, no more gravity to worry about! That means: no constraints on how you build your “spaceship” (it doesn’t need to be “something made of parts that fit inside a cylindrical rocket”). Obviously, there is mass to be displaced, but all you need is a small motor — say, a driverless ion engine tug — that can accelerate the spaceship at a constant rate, all the way. It doesn’t even need to be powerful, and that’s the cool thing about acceleration — it’s non-linear!
More than two decades ago, an acquaintance of mine — a physics engineer with some training in the aerospace industry — was challenged by ESA to design an ultra-cheap engine that could push a cargo towards Mars and back, without requiring rocket fuel and/or nuclear power.
The resulting design was a simple electric motor, powered by the sun. I remember being baffled, “how so, an electric motor? Wouldn’t that take eternities to send a craft to Mars?”
And he took a pen and paper and said, no, look, the distance between Earth and Mars is so vastly huge that what matters is not by how much you accelerate the starship, but rather that you can do it all the time at a constant rate. And he showed me how a rocket-propelled craft would accelerate the ship at, say, 1G or so, half the way to Mars, and then brake the craft (essentially turning 180º) for the rest of the voyage, until it stops at the orbit of Mars. That takes roughly six months. You could do it slightly faster by pushing the ship harder — rockets certainly can do that — but that would make the whole voyage extremely uncomfortable for any humans inside (it could, conceivably, be done for cargo-only ships).
The difference from going to 1G to 2G is, however, not going to shorten the voyage by a lot. One would expect it to take, say, half the time, but that’s not how it works 🙂
My friend’s design was, obviously, several orders of magnitude much less powerful. Let’s imagine that the best it could achieve — we’re talking about a simple electric motor spewing a few drops of water (or hydrogen gas, whatever they used for reaction that was cheap and harmless for the environment) against the bulk of a starship weighting several tons. Let’s imagine that all it can do is to afford in impulse of 0.001G. During the first days, the passengers would not travel a long way — perhaps just millimeters. They would feel motionless. After two or three weeks, they might have accomplished the fantastic feat of moving a metre or two — still seem to be stuck in the same place. But then something apparently extraordinary happens: while the spaceship seems to move very slowly at the beginning — fractions of millimetres per hour at the start of the voyage — acceleration is exponential, which means that it just takes longer to reach a certain speed, but you will reach it — assuming, of course, that you have enough space to do so. This is obviously the case on a trip to Mars.
I asked then how long it would take, using “his” electric motor, to reach Mars. And he said that the math is easy enough to do (“not rocket science” — pun intended!). Taken into account the distance to Mars, if two spacecraft were launched, one rocket-powered, the other powered with an electric motor, then the latter one would reach Mars… about two weeks later.
Of course, the longer the distance, the less the difference would be. Using the same principle to travel, say, to Alpha Centauri — and assuming that you might be able to reach 10% of the light speed — would probably take 40 years. With the electric engine? 41. Or something close to that.
Now, of course, I remember that, at the time, this was mind-boggling to me, and I protested, saying that “real” spacecraft would not accelerate all the way (or rather, accelerate until midpoint, and brake the rest of the way), because that would require gigantic reserves of fuel. “Real” spacecraft, therefore, need to accelerate fast until they reach a reasonable speed, and then maintain that speed effortlessly (with the help of some gravity-assisted maneuvers) essentially forever, without consuming any more fuel (except for slight course corrections, and that uses different engines anyway).
All that is true for conventional spacecraft, because there is an immense cost associated to producing, storing, and, most especially, carrying around all that fuel. For practical purposes, therefore, deep-space journeys follow an acceleration-constant speed-brake routine, where one expects to expend essentially zero fuel during most of the voyage.
But an electric motor with a bottle of water could accelerate all the way, without restrictions, at the same, constant rate. Granted, it would take perhaps a lot more than “one bottle of water” to reach Mars inside a 100-passenger ship. But the point is that it would be manageable, use perfectly harmless substances (you could even use sea water, or water coming from waste treatment plants — no need to waste precious fresh water), and cost little.
Well, this naturally works — it’s just a bit of math — but assumes that the ship starts its voyage in Earth’s orbit and stops in Mars’s. Thus, the space elevator. Once you’re “up there”, however, you don’t need to worry about pushing things around in the void of space — it’s easy and cheap.
A similar principle, of course, is using a light sail to capture energised solar particles — which also produce a tiny, constant force — and therefore achieve a constant rate of acceleration, without the need of carrying any bottles of water. In a sense, it’s even better, the only trouble being that, the further you’re away from the sun, the less you feel the solar wind (which can, however, be compensated by deploying larger sails). I’m also not sure how it works to brake the craft (or, well, to accelerate it towards the sun, on the return trip). Still, the Portuguese sailors figured out how to sail against the wind back in the 1400s, so I guess that a similar approach can be used.
I obviously digress 🙂 but I rather think that Elon Musk is placing his money on the wrong horse. He should invest in the space elevator instead, and then hire my friend to build an “electric engine” for him (or, well, use any other technology — there are plenty to choose from, in fact). It would possibly cost more and take even longer to accomplish his ultimate goal, but with everything in place… he would only have to pay for his investment once, and all the remaining voyages would be dirt cheap, operating at near-to-100% profit. Or, well, he could then start investing in space elevators on other planets… or on a second space elevator (built at a fraction of the cost!) simply as a backup, in the (remote) case that something goes seriously wrong with the first one (having more than two would probably never be worth the cost).
LikeLike
In terms of reaching Mars, there has been a perfectly feasible architecture for sending humans there for almost 30 years now. It’s know by a number of names – Mars Direct, Mars Semi-Direct; the Design Reference Mission (in multiple iterations) and likely one or two others. Elements of the idea have been featured in the likes of The Martian, and its cost was independently verified by NASA and ESA in 2003 ($10 billion over the first 10 years of hardware development, thereafter $1 billion a year for sustained flight operations. It’s chief advantages over other systems lay in the two-phase approach to missions, which guaranteed crews on Mars a trip home if anything went wrong with their planned Earth Return Vehicle, and also in the fact that it utilised the Sabatier reaction from the 19th century to produce propellants on Mars. Unfortunately, the proposal (1996) gained a lot of negativity within NASA, much (but not all) of which was down to a Not Invented Here (NIH) disinterest. The one part that did get adopted – the Ares launch system, which I’ve mentioned previously – actually flew its human launch system (Ares Ix) in 2009, under Constellation. Unfortunately, the NIH crowd one the day under Obama, when Constellation was largely cancelled and Ares scrapped in favour of SLS (which was essentially the same rocket as the Ares V, but required to be human-rated, making it a much more complicated beast than Ares V and, consequently proving to be far, far more costly than the combined Ares I and Ares V programme.
“Solar Sails” are interesting in that we’ve now had two successfully proof-of-concept deployments, and they are very well suited to laser propulsion – which is way more directional than solar, and with the potential to reach higher velocities – estimates being around 10% C, making them ideal for interstellar missions (such as the proposed Breakthrough Starshot). Then there are, of course both electric solar – potentially well suited to sending payloads to the outer solar system – and the Magnetic Sail (coincidentally first proposed by one of the people behind Mars Direct – Dr. Robert Zubrin).
As to the space elevator – as I’ve noted in my piece on it, the technology ain’t there – yet. And as to Musk building it – no thank you, I’ve rather have actual engineers do the work. That’s not being snide towards Musk, it’s a genuine (and objective) view of his actual engineering know-how when compared to the likes of actual propulsion and space engineers like Tom Mueller (the man actually responsible for the success of the Falcon family of rockets and also in part responsible for the success of the Dragon capsule family).
LikeLike