Why won’t Starship have an abort system? Should it?!

- Hello, it's me Tim Dodd, The Everyday Astronaut. Behind me is SpaceX's Starship prototype and something is missing. Well, it's actually missing a lot of things right now. But one thing that is missing and that there are no plans to install is a launch escape

system

or launch

abort

tower of any kind. There will be no mechanical

abort

on this rocket. Is that a good idea? In hindsight, it's always 2020, and one of the biggest criticisms of the space shuttle, which also claims to be reusable and reliable like a passenger plane, was that it didn't

have

any kind of launch abort

system

.

And that's the same kind of thing SpaceX claims about Starship. So why are we doomed to history repeating itself? And as we know, spaceX's Dragon Capsule and Boeing Starliner

have

had problems with their launch escape systems. So today we are going to debate: is all that extra material necessary? Is it really a safer vehicle? And hopefully by the end of this video we'll know whether or not it's a good idea to put people on top of this rocket without any kind of mechanical abort system. Let us begin. - Three, two, one, take off. (upbeat music) - That's one small step for man. - Subject to the test zone. - This is another one of those topics I've been asked about time and time again and for good reason!

Especially now that we are seeing the commercial crew program put launch abort systems to the test. And this has become even more relevant since we witnessed the failure of the MK-1 spacecraft prototype in Boca Chica while performing pressure tests on the launch pad. What if there were people on board? Why isn't SpaceX thinking about putting a launch abort system on it? As you know, I've already talked quite a bit about abort systems in a video explaining why both SpaceX's Crew Dragon capsule and Boeing's Starliner have opted for liquid-fueled thrust abort systems, rather than a thrust system. more traditional with solid rocket motors. .

So if you need an overview of those systems, definitely watch that video! It

should

help you gain additional perspective on modern abortion systems. Now, in case you didn't notice when you clicked on this video, this is another pretty long video. As you know, I don't like to skim the surface of topics, I like to dig deeper and delve into the data and history to find the answers. So let's get into a lot of little details, graphs and data and more graphs and more data. We'll go over rocket certifications, the reliability of rocket engines, the risks and benefits of Starship designs, and we'll even look at the entire history of human spaceflight and discover how many times an abort system has actually saved lives. of those who are on the board.

At the end of this video, it won't just be my opinion on whether Starship without an abort system is a good idea or not, it will mostly be an analytical summary. So to make it easier for your viewing pleasure, here are the timestamps for each topic we're going to talk about in case you want to go back and review something. There are also links in the description below to those timestamps and an article version of this video so you can easily review some numbers and sources. (upbeat music) Okay, then abort the systems. The idea is quite simple.

Rockets are fickle beasts. They are following the best possible engineering line. They must withstand extremely intense loads, temperatures and environments while being as light as possible. Since there are literally millions of moving parts, it's honestly a miracle that they work. So when you put someone on top of a vehicle, it's basically a giant pump with a nozzle that has to have millions of parts working properly to not fail. It is generally considered a good idea to have a backup if things don't work out. It's not going according to plan. Throughout most of the history of human spaceflight, there has been one method that has dominated launch abort systems: a solid rocket motor launch escape tower.

This is the tower above the crew module that you may have noticed on vehicles like Mercury, Soyuz, Apollo, and even NASA's Orion capsule. Some vehicles like Vostok, Gemini, and the early space shuttles had ejection seats as an escape system, and I have a video on that if you want to know more about it, but regardless of whether it's an ejection seat, a launch tower, or a More modern liquid push system, the concept is the same. If the exhaust system detects a loss of thrust, a severe deviation in flight path, or even detects a rupture in the tanks, it will activate the abort system.

Early rockets, and probably some modern ones, I couldn't really find the answer to that, have a trio of sensing cables along the fuselage. If two of those wires lose contact, this indicates that the fuselage has broken, which would then activate the abort system. The launch abort system can also be activated manually by an astronaut who may notice a critical defect that the system does not detect. Once an abort system is activated, the stage separation system must release the spacecraft while the abort engines ignite, in order to remove it from the rocket. When the engines actually ignite, they accelerate the crew capsule and move away extremely quickly.

Abort engines can generate some serious G's, such as over 10 G's. Therefore, launch abort systems were considered pretty much the best way to keep the crew safe. But what if you made your entire system more secure? What if you created more redundant systems and oversized each part to achieve even greater safety margins? (upbeat music) Designing a rocket to be as reliable as a passenger plane, that's exactly what NASA hoped to do when designing the space shuttle. - The vehicle was called the space shuttle, a spacecraft similar to a passenger plane capable of making more than 100 trips from Earth to Earth orbit. - The idea is that if every part of the space shuttle were over-engineered, the possibility of loss of the vehicle and crew would be very low.

So low that it might actually be safer than adding the additional moving parts and systems needed to create an effective abort system. During the Apollo era, when NASA had what now seems like unlimited funding, they hired General Electric to do complete numerical probabilistics, how do you say that word? Probabilistic assessment of the risk of taking a human being to the Moon and returning him safely to Earth. The number GE got was five, 5%. 5% chance of successfully landing a human being on the Moon and bringing them home. NASA Administrator James Webb didn't like those numbers, and rather than make changes to the rocket and the mission, he decided to change the way they designed around risk.

Which, let me point out, is not necessarily a bad thing. They developed something called Failure Modes and Effects Analysis, which was a way to identify designs and hardware that, in the worst case, would lead to a catastrophe. These were classified as Criticality one, which would threaten the life of the crew or the existence of the vehicle. Criticality two that would threaten the mission or Criticality three for anything else. They also added an R as a notation for redundant systems in these parts and design analysis. With NASA's budget beginning to shrink once humans landed on the Moon, and Congressional support for the next Shuttle Program waning, NASA had to sell the Space Shuttle as a cheap and reliable workhorse.

And even their exact risk had to be calculated to launch plutonium-fueled spacecraft like the Jupiter Galileo probe. Testing the reliability of a complex system involves studying the system as a whole, identifying potential failure points, and then gathering the boundaries of these potential failure points under different environmental conditions. Today, it also tests them in statistical models and computer simulations, and iterates on them to ensure that the computer simulations meet real-world performance to help determine whether the probability of success is above the required threshold. Each and every part will have a design specification and a certain safety factor on how much more it can support beyond its design specification.

The rule of thumb for most rocket parts is a safety factor of 1.5. All of this really means that if a part needs to withstand 10 newtons of force, when tested it

should

be able to withstand 15 newtons of force without failure. But when there is no good test or heritage data for a part, a safety factor of 2.0 is considered a good rule of thumb. In other words, designed to handle twice as much as necessary. But at the beginning of the Space Shuttle program, estimates of catastrophic failures ranged from less than 1% to less than 0.001%. So a couple of orders of magnitude into your estimated range of predicted safety.

Despite NASA's optimistic nature, the first four space shuttle flights had active ejection seats, but they were later abandoned due to their very limited use cases and the fact that they could only eject the crew from the upper deck as The rest of the crew was on the center deck below them. After the Challenger disaster, NASA actually considered using an ejection cabin that could eject the entire Shuttle crew module, like the F-11 and early B-1 prototypes, but it was considered too complex, too heavy, and too necessary. many modifications to make it really feasible. So in retrospect we know that the shuttle ended up with two failures out of 135 flights, one during launch and one during re-entry, giving it a success rate of only 98.5%, which is a long way off even the most conservative safety estimates.

Likely due to the space shuttle's history, NASA changed its certification system for the Commercial Crew Program, which requires having a crew loss probability of 1:500 on ascent, 1:500 on descent, with a probability of 1: 270. of a problem while in orbit. To truly certify and validate systems, sometimes it is simply a matter of putting everything through its paces to test the system as a whole in addition to testing each individual part. Testing the system as a whole is called total testing, as NASA coined it during the Apollo era. In the long term, full testing could be a faster route to verifying the system rather than acquiring certification through more ground testing and analysis. and use heritage data that may require a higher safety factor per piece if it is not to be fully tested.

You can see this difference in methods being developed today in how SpaceX chose to validate and certify its abort system by opting to test it in flight, while Boeing chose to certify its abort system through more stringent parts certification. individual. Since the space shuttle and spacecraft lack abort systems, what design considerations did the space shuttle have that were so dangerous? And does Starship share any of the same flaws? (upbeat music) There are a few things that made the space shuttle inherently dangerous, but let's start with those solid rocket boosters. Those gigantic white solid rocket boosters on the side of the external fuel tank actually provided more than 60% of the thrust at liftoff.

Here's the thing, once they were turned on, there was no way to turn them off, you go somewhere and you go somewhere in a hurry. Hopefully the pointed end is facing up and the flame-shaped end is facing down. This meant that any abort mode or procedure, no matter how bad or terrible, required the crew to override the thrusters, seriously. So for a full 127 seconds, if there was a problem, you just had to cross your fingers and toes that it wasn't too serious and you could move on. This is mainly because if you aborted by disconnecting the orbiter from the external fuel tank or attempting to jump, you would end up inside the plume of those SRBs, which would be very, very lethal.

Unfortunately, some problems cannot simply be fixed, such as on January 28, 1986, when the space shuttle Challenger lifted off on its tenth mission. A leaking O-ring sealing one of the sections of the Solid Rocket Boosters generated a leak that ended up causing the gasket holding the SRB to the external fuel tank to separate, which then led to a complete structural failure of the vehicle and a tragic loss. of the crew of seven. But perhaps the biggest problem with the Challenger disaster was not a hardware problem, but a problem with program management and the pressure to get that particular flight off the ground.

It was known that they would be launched outside the SRBs' predetermined operating framework and it was recommended not to launch that day.Although using liquid rocket boosters probably wouldn't eliminate all potential failures, and could arguably be less reliable than a solid rocket booster, liquid rocket boosters can at least be shut down, which can generally open up more options. of abortion. Take a look at this graph. Notice the black sections. Yes, those are climbing sections where there would be a total loss of control and/or structural failure if the vehicle lost two or three main engines. Basically, you lose two of the three main engines and you're screwed.

And you're not even talking about having any problems with SRBs or anything else. But after the Challenger disaster, NASA came up with many more contingencies, including alternate runways and the iconic orange advanced crew escape suits to, you know, escape if things went wrong. Literally escape. Here are the new abortion profiles. See these gray sections during the climb? Now the plan was for the crew to literally jump during those sections. As in, carefully get rid of the external fuel tank, make the orbiter glide stably, literally unzip it, blow up the hatch, extend a damn pole that made it so you wouldn't hit the wing when you jump, and then jump, seriously .

Speaking of hitting the wing, that's another major flaw of the space shuttle. The orbiter was hanging off the side of the vehicle, putting the crew module and fragile thermal protection system directly in the path of ice and foam hits. The orange external fuel tank housed cryogenic hydrogen and oxygen that had a good amount of insulation to keep them at operating temperatures. In fact, sheets of ice can be seen falling from basically any liquid fuel rocket during takeoff, it's a known variable, and large chunks of foam were even observed falling from the space shuttle's external fuel tank, but NASA accepted every time plus this fact.

When you combine a large chunk of foam and the fragile nature of the space shuttle's thermal protection system, you have real potential for disaster. And this is exactly what caused Columbia to fail. A large piece of foam hit the leading edge of Columbia's wing during climb, tearing a large hole in the reinforced carbon-carbon heat shield section of the left wing. The shuttle and the seven crew members on board continued to fulfill their mission for 15 days with a hole that would condemn them to re-entry. On February 1, 2003, as Columbia re-entered Earth's atmosphere, the large hole in the wing caused hot plasma to essentially destroy the wing and therefore the orbiter, tragically ending the lives of the two. seven crew members who were on board.

In addition to the orbiter being thrown off the side of the rocket and endangered by falling debris, the danger was amplified by how fragile the space shuttle's thermal protection system was. The space shuttle's more than 24,000 silica tiles were literally glued to the orbiter's aluminum airframe that covered the entire underside of the vehicle. Their fragile nature not only caused many headaches, but also led to some difficult times. Perhaps the most notable was STS-27, which experienced a debris impact 85 seconds into the flight. This knocked down one tile and damaged OVER 700 more tiles! Oh! In the most fortunate of circumstances, the missing tile was right on top of a steel mounting plate for the L-band antenna, the steel of which has a higher melting point than the aluminum fuselage and, by sheer luck, the orbiter survived re-entry and did not.

It won't end in a disaster like Columbia would some 15 years later. (upbeat music) Now that we know some of the major design flaws that plagued the space shuttle, let's review Starship and see how its design is different. Well, right off the bat, Starship is on top of the Super Heavy booster and not hanging off the side. As we've noted, this is clearly a safer place for the crew cabin, putting it ahead of potential debris impacts. But we must point out something that will definitely need to be studied. Remember, Starship also has those big flaps that are vital for your reentry.

This means there is a chance that ice from the liquid oxygen and liquid methane tanks on Starship's upper stage could hit the leading edge of the fins. This is where a stainless steel body and the fact that the heat protection plates will be bolted on has a big advantage over the fragile silica plates and reinforced carbon-carbon that covers an aluminum frame. So let's not forget that Starship's leading edge and underside of the fins will still use a heat shield, but unlike the space shuttle's silica slabs, they are supposedly much more durable and are actually bolted to the fuselage instead. of sticking.

We've already seen SpaceX test heat shield mounting and material options on Starhopper, subjecting them to extreme environments, vibrations, and temperatures. They also experienced reentry when SpaceX placed some small pieces of Starship's heat shield into a Dragon capsule to observe reentry. Steel body panels and some variant of a TUFROC heat shield bolted to the airframe should be stronger than the space shuttle, considering that steel can ding and dent and won't break completely like reinforced carbon-carbon or it will just fall off like silica tiles. But also, like STS-27's lucky brush with death, steel has a much higher melting point than aluminum.

This allows for much greater heat tolerances than the aluminum structure of the shuttle, as we have discussed. In fact, Starship can likely survive almost completely intact with little to no additional thermal protection, as other stainless steel components have survived reentry virtually completely intact. Now I already made a video, because it even talked about purging fuel, to do a form of transpirational cooling, but that fell off the table or could be used in some critical points. This is one of the main reasons why SpaceX switched to full carbon fiber because when you take into account the size of the heat shield needed, stainless steel begins to take advantage as it can withstand much higher temperatures overall sooner. to start to fail.

Well, hopefully a stainless steel body and a stronger bolt-on thermal protection system, along with placing Starship at the top of the rocket stack, should help mitigate the risk of a reentry disaster, like Columbia. So how about we put all of this on top of a rocket booster that will have at least 37 of the most advanced and complicated engines in the world, which is, of course, the full-flow staged combustion cycle Raptor engine. How could this be safer and have fewer failures than a simple pair of solid rocket boosters that have virtually no moving parts? This is where SpaceX has good knowledge and experience.

Their Falcon 9 does something quite unusual in the rocket world, which is to have nine smaller engines on the booster instead of one or two larger ones. This actually allows for multiple motor output capabilities. Now, depending on which engines fail and at what point in the flight they fail, nine engines give the Falcon 9 an additional safety margin compared to other rockets. Each engine is isolated within a blast containment cell as part of the Octoweb configuration. That's why you can have one engine failing and not affect other engines. Combine this with modern sensors and fast-acting computers, and the rocket should be able to shut down an engine before that engine suffers catastrophic failure.

SpaceX has developed its Merlin engine to the point of achieving extreme reliability. In fact, only one Merlin has stopped in flight of the more than 800 that have flown to date. Not to mention, it was very early in the program, the fourth flight of a Falcon 9 for the CRS-1 mission to be exact. And since then the engine has been 100% reliable. So in total the Merlin engine to date has proven to be 99.88% reliable, if we combine that with redundancy in the first stage we get an incredibly reliable powerplant. Not to mention the fact that some of the Merlins in the Falcon 9 first stage have to fire two or three more times per flight in order to land.

But just for fun, let's take a look at other rocket engines throughout history and see how reliable they have been. But just a little warning, it gets really hard to just say reliability, because there are a lot of other factors, especially when you take into account restarting the engines, or running out of TEA-TEB ignition fluid or something like that. But still, looking at these numbers should give us a decent perspective on how reliable liquid fuel rocket engines can really be. Again, SpaceX's Merlin engine currently has 99.88% reliability in flight, making it slightly more reliable than the RS-25 space shuttle main engine, which also shut down only once in flight, but with only three engines in 135 missions, finished. with 99.75% reliability in flight.

Then there's RD-180, which technically shut down four seconds early on one of its 86 flights to date, making it 98.83% reliable, although it's also 100% reliable since that particular mission, OA-6. was able to continue. and it will still be a success, but just barely. So we can say between 98.83% and 100% depending on how motor reliability is defined. I mean, shutting down early is considered engine failure; if that had happened any time before, the mission would have failed. But I would consider it a successful mission for the Atlas V as the Centaur upper stage could make up for the failure. But how about another Russian engine?

The RD-107 and its brother the RD-108 that propel the Soyuz rocket. Now, this rocket has flown so many times and for so long in so many different forms, that it's definitely not fair to compare the early days of the Soyuz rocket, plus the data is really hard to find. So let's look at the 267 flights in the 21st century using an RD-107 and an RD-108, of the 1,335 engines fired in this century, only one has failed, giving the RD-107 and RD-108 a 99.92 % Of flight. reliability. But one of the most reliable engines ever flown was actually the monstrous F-1 engine that powered the Saturn V.

The 13 times the Saturn V flew, the 65 F-1 engines that powered those flights were 100% successful. Now before you jump into the comments section and say, didn't Apollo 13 miss a core F-1? No, that was a J-2 in the second stage. There seems to be a Mandela effect in F-1 reliability where everyone, including me, tends to think that an F-1 failed the climb. So let's assume that since there will be a lot of data on the Raptor engines, since Starship has a lot of them, and they fly Starship a few times over and over again, SpaceX will eventually match the reliability of the Merlin engine with multiple engines. engine capacity, the booster should be a very reliable first stage.

Assuming SpaceX does its due diligence to prevent an engine failure from affecting other nearby engines like they did with the Falcon 9, and how the N-1 didn't do a very good job of that. Having dozens of engines can make it an incredibly robust and reliable vehicle. Okay, sure, having dozens of engines on the rise could help make the booster safe and reliable. But what about the big elephant in the room? For humans to survive a trip on Starship, the Starship itself must not only perform a rather extravagant landing maneuver, but must also rely on two of the three working Raptor engines for landing.

Is that really safe? Can we really rely on propulsive landing to save human lives? Well, let's go ahead and take a look at the Falcon 9 again, since it's one of only two vehicles in history to make propulsive landings after reaching space, but remember, it also doesn't reach orbital speed, like Starship does, so We're just going to look at the landing aspect, just the landing lights up and we'll use engines as propulsive landings for now. Because the end of Starship's proposed landing process will look and function almost exactly like that of the Falcon 9. To date, SpaceX has made 46 of 54 landing attempts.

That doesn't sound very good, but remember, before the first landing, it was considered literally impossible. No way, when it happened it was a big deal. If we just look at the landing attempts that started in 2017, after it became a little less experimental and they figured it out, we get some pretty surprising numbers. There have been 45 attempts since 2017 and only three of them were failed landings. None of these three landing failures were due to a failure of the Merlin engine itself, although again that gets complicated. Inorder, on February 6, 2018, SpaceX launched the first Falcon Heavy on its demonstration mission, landing all but the central core of the Falcon Heavy.

The central core was out of TEA-TEB, which is the pyrophoric starting fluid that actually starts the engines. It seems like there was a pretty easy solution and it wasn't really that they would need more TEA-TEB, but that they just needed to change which bottle they got the TEA-TEB from and at what point to solve the problem. Later that same year, on December 5, 2018, for the CRS-16 mission, a new Block 5 Falcon 9 core failed to land when the hydraulic system that controls the grid fins became stuck. The solution was simply a purge valve that would prevent this from happening again.

As far as the Merlin engine was concerned, it worked perfectly well and allowed the vehicle to land so smoothly that it did not break up and could be towed back to port. Lastly, the last Falcon Heavy launch on June 25, 2019 for the STP-2 mission also had a failed landing attempt on the central core. SpaceX did not expect the core to survive reentry due to the extremely high speeds and pushing the vehicle to absolute limits. As far as we know, the engines still fired and ran fine, but the thrust vector control that directs them was destroyed by that pungent reentry heat, causing the vehicle to lack precise control for landing.

So the motors themselves were fine, but the TVC was a bit destroyed, so it's hard to say. Since this was at the outer limits of what the booster is capable of, the solution is to simply not push it as hard if you need to land. So if this were a Starship mission with people on board, they would have made sure there were healthy margins that were safely within Starship's operational range and either wouldn't even accept a mission or design a mission where they would push it to the limit in the first place. So can we ever rely on propulsive landings for humans?

In the end, for sure. Think about it, the Apollo missions were based on supposedly landing on the moon, and that worked out pretty well. As for daily use, as long as there are redundancies. Having three engines on and having engine off capability is a good starting point. But what about other systems installed on Starship? What's with those giant things with wings and fins? What happens if the hydraulic system fails and they get stuck, as happened in CRS-16? Well, for this we need look no further than airliners and the space shuttle. This is where the old redundancy comes into play.

Planes would lose control and not be able to operate their landing gear as planned if the hydraulic systems failed. The same goes for the space shuttle. Which is exactly why there are redundant generators, redundant pumps, redundant lines, basically everything in that system is redundant. It's actually not very smart to compare the CRS-16's failed hydraulic system and say, see, what if that happens on Starship? Because landing a Falcon 9 booster is not mission critical, much less necessary for human safety, they have intentionally lacked redundancy for simplicity. But there is one big thing that is not redundant and can be catastrophic if it fails, and boy, do I mean big, because that's the fuel tank cutting into the fuselage.

Honestly, this is my biggest area of ​​concern. And to be honest, it's the only thing SpaceX has had bad luck with time and time again. The first Falcon 9 failure was due to a helium-filled composite-wrapped pressure vessel, or COPV, these pressurize the fuel and oxygen tanks, and one became loose in the upper stage oxygen tank on June 28. 2015 for CRS-7. mission. This led to a rapid unscheduled disassembly and total loss of the rocket and its payload. The next failure was the infamous AMOS-6 anomaly on September 1, 2016. Again, overpressurization due to a helium tank failure in the upper stage caused a complete and total loss of the vehicle while refueling on the launch pad. launch. for a static fire test.

Then we have the Crew Dragon anomaly on April 20, 2019. This was when SpaceX was testing the ground launch abort system and a frozen chunk of nitrogen tetroxide shot through a titanium valve, causing a rupture in the system and total loss of the vehicle. And more recently, and perhaps most alarmingly, we saw a Starship MK-1 prototype fail on November 20, 2019 when we saw it blow up due to overpressurization during a pressure test. Now I'll give this a little pass. This vehicle and that test were nowhere near future operating conditions. SpaceX was pushing this prototype well beyond normal operating ranges and these were the jagged welds of the initial prototype and do not represent a future, or more refined vehicle as it will be.

But it sure gets a little creepy when you think that if this was a fully fueled Starship and there were people on board, there would have been no chance of aborting it. (upbeat music) I guess this brings us to the question of what options does Starship have for aborting? We've already gone over what design considerations allow it to avoid the same design flaws as the space shuttle, but it still lacks an actual abort system. So can you have an abortion? Well, first let's make sure we are clear about the types of cancellation options. There is a big difference between a platform abort, an in-flight abort, an in-orbit abort, and a mission abort in general.

A pad abort is the option to release the spacecraft from the rocket while it is still on the pad. This is actually a quite dangerous time while the vehicle is full of fuel and filled with highly pressurized explosive material. So, can Starship abort the platform? Yes and no. If the problem is in the upper stage of Starship, like a tank rupture or something, the simple answer is no. But what if the problem is in the reinforcement? If the booster suddenly breaks down, Starship's upper stage could perform an emergency jump start of the Raptor engines, which could help prevent Starship from simply falling into a pile of what was once super-heavy booster and is now a hellish landscape in flames.

If all engines are on, including the vacuum-optimized engines, Starship would barely have enough thrust to slowly move away from the pad and divert to a safe landing zone. That assumes that a failed Super Heavy booster didn't damage the Starship enough for her to be unable to fly. Don't forget, you're not trying to escape the pressure wave, because spoiler, unless you can get to zero at the speed of sound literally instantaneously, you won't be able to escape it. So if you're inside a pressure vessel, you should at least survive the initial explosion. You might also wonder how they could power vacuum-optimized engines at sea level.

That was a big part of the air spec, you can't really do that. Well, according to Elon, they could have a double-bell mouthpiece design and attach the mouthpiece to the helmet wall to stabilize it. And in general, yes, you can actually shoot a standard vacuum nozzle at sea level in an emergency, but it will very likely fail, but why not try it if the other option is a complete failure anyway ? So, the pad abort, a little, maybe a little, well, at least it's better than the space shuttle's complete lack of abort options while sitting on the pad.

The same goes for an in-flight abortion. An in-flight abort is exactly what it sounds like: aborting while the rocket is flying. Again, assuming the upper stage is not to blame for the problem, the spacecraft could theoretically move away and perform any maneuvers necessary to re-enter and land elsewhere or, in the worst case, perform a soft emergency landing. Again, it should have more options and opportunities than the space shuttle. Once Starship has separated from Super Heavy, there are really no abort options other than aborting to a safe reentry profile and then re-entering if it can't reach orbit.

But fortunately, with Starship's control surfaces, they could greatly alter its aerofoil and dynamics for a handful of safe reentry options. This is useful compared to, for example, a capsule that cannot change its shape and actually has areas in a launch profile that engineers must avoid. For example, if a standard single-engine RL-10 Centaur upper stage were used on Boeing's Starliner, there would be large portions of the profile where an abort would be fatal due to extreme reentry forces. Boeing and ULA therefore had to opt for a twin-engine variant of the Centaur upper stage that could fly with a safer profile that allowed a safe abort window throughout the climb.

And, of course, upon re-entry there is never a cancellation option. Reentry simply needs to work. Even if there was a mechanical cancellation option during re-entry, it probably wouldn't help much. Now, of course, a passively stable capsule with a simple ablative heat shield has very little that can go wrong, but again, redundant hydraulics for the control surfaces and a structure that can withstand high temperatures overall should offer a decent cushion. on Starship re-entry. (upbeat music) Well, we're finally getting to the real heart of the question. Is it really better to have an abort system on a rocket, period?

To do that, let's quickly go through all the aborts and all the human spaceflight accidents and determine whether or not an abort system could have helped. If we look at the history of orbital spaceflight, only 18 deaths have occurred during orbital spaceflight activities. The first death was a parachute failure on the first Soyuz mission in 1967, in which cosmonaut Vladimir Komarov died. An abortion system would not have helped. The next tragedy was Soyuz 11 in 1971, whose ship decompression caused the death of three cosmonauts. To this day, this is the only incident in which humans died in space above the Karman line.

An abortion system would not have helped. Then we have the space shuttle Challenger disaster in 1986, which we've already talked about. A mechanical abort system likely would have saved the crew of seven. Lastly, we have the space shuttle Columbia disaster in 2003, which again tragically killed all seven crew members. An abortion system probably wouldn't have helped. There's a chance that if an escape pod had its own heat shield and the like, it might have helped, but aborting during reentry is unlikely to be a very good option. Now let's look at the number of times a manned orbital launch escape system has aborted.

This number is very small. Today, an abort system actually armed for a flight has only been activated three times. The first time an abort system was used was actually on an uncrewed Soyuz test flight for the first Soyuz 7K-OK mission in 1966. The launch was restarted when a strap-on booster failed to ignite. The ground crew went out to inspect the rocket when suddenly, 27 minutes into the scrubbing, the launch tower activated because its gyroscope noticed that it was 8 degrees off its axis from where it thought it should be due to the Earth's rotation. . The abort system firing ended up setting the third stage on fire, and then the rest of the rocket exploded on the pad, killing one ground crew member.

In this case an abort system caused a failure and a death. The next time an abort system was activated, it was the only time there was an abort on the platform with a crew on board. On September 26, 1983, when the crew of Soyuz T-10-1 had to abort their Soyuz rocket that had caught fire while still on the launch pad. After aborting and landing safely four kilometers below, the crew was bruised and shaken when they were met by the recovery team. They gave them cigarettes and shots of vodka to relax. In this case, obviously, the launch abort system saved lives.

Lastly, the abort system was activated on another Soyuz MS-10 mission in 2018 when there was a booster separation issue that caused one booster to break off the core stage. This activated the abort system, not the entire turret that had just been jettisoned a few seconds earlier, but a smaller abort system built into the fairing covering the crew module. In this case, the abort system saved lives, but perhaps simply shutting down the engines and disconnecting from the booster would have been enough without an active mechanical escape system. In reality, only two other flights were aborted, the first being Soyuz 7K-T-39 in 1975, which was aborted after the escape tower and fairing were discarded.

A cancellation system would obviously nothelped, since it had already been abandoned. . Then there's the only space shuttle to abort, STS-51-F in 1985. It performed an in-orbit abort maneuver when one of the RS-25 main engines shut down. Again, an abortion system would not have helped because it was unnecessary. So in the grand scheme of things, to date, a mechanical abortion system has only saved lives twice, may have prevented a tragedy, and in one case actually created a tragedy. So, of the 320 crewed orbital flights to date, only three missions in total required the use of an abort system, or less than 1% of crewed launches.

There were three other launches where an abort system would not have helped at all, and two that aborted without any type of escape system. And if you look at the last 30 years of human spaceflight, from the '90s onwards, only one of the 180 launches required the use of a launch abort system, so only about half a percent of flights would get any benefit from a release of the exhaust system at all. (upbeat music) Now, before we answer the question of whether abortion systems are necessary, let's look at one more thing. How can we improve overall rocket safety so that we don't need a period of system outage?

I think the answer to this question is that we need to fly more, much more. And we need to reuse systems over and over again so we can see where they are weakest and where we can make the biggest security improvements. Let's take a look at airliner safety. This is a graph showing how many miles of commercial airline travel occur by accident over time. Unfortunately, this data only goes back to 1929 and doesn't even show the early days of the Wild West of air travel. But in less than a century, the industry's safety record has improved by three orders of magnitude.

Now I really wish we had data on the first three decades of human air flight, but unfortunately the data is not available, but I wouldn't be too surprised if it didn't look very far off from this graph. This is actually the success rate of orbital launches per year. Notice how quickly humans reached the top 90th percentile. But then it stalled. Let's compare that to the airline industry during the same time period. Yes, by then humans had already learned how to fly, it's not until you get into the tens of thousands that you can even begin to decipher an improvement in flight success rate because we're already chasing the nines in reliability at this point. spot.

And I think there are a few reasons for this. First of all, in total there have not even been 6,000 orbital launches. Compare that to the nearly 40 million commercial air departures in 2018 alone and you'll realize how rare spaceflight remains. I bet the first 6,000 attempted airplane flights also had an equally terrible flight record. 6,000 flights were probably achieved in a much, much faster period of time and with a significantly lower barrier to entry than building a rocket. Now, we probably shouldn't compare rockets to airplanes, because getting a rocket into space and bringing it back safely is really, really hard, orders of magnitude harder than flying a plane to begin with.

You can make airplanes out of paper, or if you're Peter Steeple, you can make airplanes out of just about anything. But it's fun to see how quickly we can improve reliability, it just takes longer and longer. But it's fun to compare how quickly things can become reliable once you do them enough. The answer to what we can do to make rockets more reliable is simple. Fly more frequently and fly reusable rockets over and over instead of throwing them away. Only then will we begin to approach the reliability of an airliner. (upbeat music) So it's time we finally put this all to rest.

Launch abortion systems. They are needed? Do they really make astronauts safer? Do we need them in the future? If so, will we always need them in the future? So remember when we looked at how many times a launch abort system would have saved the crew's life and it's a surprisingly small number? Well, I still think it's a good idea for this generation of rockets. I think NASA, SpaceX, and Boeing are correct in assuming that the Falcon 9 and Atlas V, as reliable as the rockets are, still lack significant flight data to really be considered safe enough without a launch abort system. .

But don't forget that the launch interruption system still brings with it its own complications and problems. Remember how SpaceX's Crew Dragon capsule exploded while testing the abort system? And in fact, Boeing also had problems with its launch abort system catching fire on the pad. Basically, you're taking more parts and a small rocket and connecting all these additional systems, which can also fail, and you're putting them directly into your crew module anyway. Sure, a lot of work is being done to make them safe, but you're solving the rocket problems by putting more rockets on them. That would be like putting a Cessna propeller plane inside a 747, in case the 747 fails you can fly the Cessna.

It's probably better to make sure the maintenance is up to date than buying a Cessna. It's like how people ask all the time if Super Draco Abort engines could be used as a backup for parachutes if they fail, and the answer is, technically, yes, of course they could. But by the time you certify that procedure, those systems, and make them safe and reliable, it probably would have been better to make sure your parachutes didn't fail in the first place and make them more reliable. Likewise, would you rather design an abort system, bring all of these procedures, envelopes and subsystems together, or focus on making the entire vehicle much safer?

At some point, an overall more reliable system can arguably be achieved by having fewer parts. I think Elon Musk is right when he said this in the 2019 Starship update: the best part is not participating. The best process is no process, it weighs nothing, costs nothing and can't go wrong. As obvious as it may seem, the best part is not participating. What impresses me most when I have design meetings at SpaceX is what did you undo? Undesign is the best. Just delete it, that's best. - So I guess the question should be: would you travel in a spaceship without an abort system?

For now, the answer is no. I think we should see at least a few dozen uncrewed flights first. We should find the limits and boundaries, maybe have a few glitches or two and only once we've seen Starships fly 10+ times reliably without glitches would I consider getting on one. But I'm also a chicken, I don't think I'm cut out for anything less. I think we may see humans willing to take a chance on Starship early in the program, and if there are NASA astronauts on board, I wouldn't be surprised if they needed an abort system. Especially since SpaceX will likely load and deliver its fuel, just as they do with the Falcon 9, meaning the crew will need to be on board as the fuel flows.

In general, it is more dangerous to fill and pressurize a vehicle than to have it there stable and full of fuel. So unless SpaceX can change that procedure for Starship, I honestly can't imagine NASA would want any of its astronauts back on board without an abort system anytime soon. At the end of the day you can't solve problems you don't know exist. Just as SpaceX was so surprised to discover problems with cryogenics and overwrapped composite pressure vessels, or a strut failure or a titanium valve exploding, sometimes a design flaw simply isn't discovered until it rears its ugly head. head.

That's why I think it's vital that we see these things fly, fly often, and fly again and again. Only then will I think there is a track record of proven reliability that would make it a safe enough option to not have an abort system. So what do you think? Abortion systems, are they good or bad? Would you get on a rocket that doesn't have an abort system? At the end of the day, that's what it really comes down to. I can see both sides of this argument. I understand you could say it can obviously make things safer.

Because if the rocket fails, at least you have a backup rocket, while also adding more parts. So maybe there's some elegance in the simplicity of the design that doesn't have all these extra parts and just focuses on making them more reliable. But I do not know. Let me know your opinion in the comments below. And also be sure to let me know if you have any other questions about Starship, abort systems, stainless steel, Raptor engines, space shuttles, anything. Let me know if you have any questions and I'll be sure and try to include them in my video list.

I have this endless list of videos that I just can't keep up with, but I'm so excited to keep making them. So stay tuned, there's a lot of really good content coming your way. As always, I owe the world's biggest thanks to my Patreon supporters, who literally helped me do all of this. They've been a huge help on this video, trying to find all these fun facts and finding little bits of information with me, crunching the numbers and confirming that my stuff is accurate. So if you want to help do what I do, consider becoming a Patreon supporter by visiting Patreon.com/everydayastronaut, where you'll get access to our exclusive Discord, Patreon, and exclusive live streams.

Thank you. And while you're online, be sure to check out dailyastronaut.com/shop for limited edition products. Now notice the merchandise is getting amazing, they are all hand sewn and screen printed patches. Custom neck labels, all this really cool stuff. And these amazing designs are limited runs. So if you like something, you better get it now, because there's a good chance it won't be there the next time you click on the website. And I guess you can even grab what is apparently one of Elon Musk's favorite t-shirts, the Full Flow Stage Combustion Cycle T-shirt. We also have it in a hooded version, just in time for the cold weather.

So come in while you can and get some amazing products for yourself. That's dailyastronaut.com/shop. Thank you all, that will be enough for me. I'm Tim Dodd, The Everyday Astronaut, bringing space to Earth for everyday people. (upbeat music)