Elon Musk's New Version Starship Is Actually Crazy

We've had three V-3 flights. So, what's the most prominent problem? Well, it's clearly the engine issue in the Super Heavy. Reliability remains a big question. So how do we address those problems? That's where the V-4 might be the next step to further improve Starship. Signs of such a version are emerging. What makes it special? Let's find out. It could be said that, by some coincidence, SpaceX built the Starship program with each year representing a different version. 2023 and 2024 were the years of V1, with six flights. 2025 was the year of V2, with five flights. And 2026 was the year of V3. And there may be many more exciting things to come. So, you should definitely subscribe if you haven't already, and then visit the channel regularly for continuous updates on SpaceX and the aerospace industry. Of the versions, V3 represents a major change in the Super Heavy and ship, addressing the problems of the two previous versions. The problems in V1 were more in the ship, while in V2 they were in both stages. A key highlight is that both versions have booster landings using chopsticks, and ship has begun its return journey, although the results haven't been very successful. By V3, ship has seen the biggest improvement. Not only was it able to return in a controlled manner, but it also had significant improvements in its heat shield, allowing it to reach orbit and deploy its first operational payloads on Flight 14. So what about the Super Heavy? Well, it had changes to the hot staging, grid fins, and engines. The engine was upgraded from Raptor 2 to Raptor 3, a much simpler version. Therefore, it was expected to improve reliability. But let's look at the reality. The first three V-3 and Super Heavy flights all had engine problems. Flights 12 and 14 both had one engine completely fail during takeoff. This often leads to subsequent consequences, such as further engine failure, erratic boost-back burn, delayed landing burn, or insufficient deceleration performance. As a result, perhaps only flight 14 had a smooth landing, while the other two were hard splashdowns, with boosters exploding before impact, and landing speeds remaining very high near the destination. Okay, that's the issue with Super Heavy V3 that I've placed in the Starship's historical timeline. So how will SpaceX handle it? Of course, SpaceX will still try to improve with each flight. This is considered a fundamental upgrade. But besides that, a more groundbreaking and long-term solution is to upgrade the version. And the next version will definitely be V4. This version has actually been mentioned by Elon Musk from last year to this year. One of the notable upgrades is the addition of two more engines to the Super Heavy. This will increase the number of engines in the Super Heavy to 35. Not just in theory, SpaceX seems to have already prepared the hardware for that version. Some sources reveal that SpaceX is preparing something at Starbase, including pins showing 10 pins for middle ring engines, and possibly up to 5 inner ring engines. Unfortunately, I haven't seen any precise images of this upgrade. Yeah, the increase in engines on the Super Heavy is precisely adding two more engines to the inner ring. What impact does this have? Remember, the inner ring engines have gimbal capabilities, so they play a much more significant role in a flight. Besides providing thrust, they also provide steering. Therefore, they operate almost throughout the entire flight. If the ascent involves all 33 engines, then in MECO, while the outer and middle rings are gradually shut off, the inner ring engines are kept running to maintain thrust for the Super Heavy during separation. Then, during boost backburn or landing burn, the inner ring engines are always active. To date, with a three-engine configuration, the inner ring engines still perform well. Landing issues or boost backburn mainly stem from the middle ring. Therefore, in addition to a direct fix, the SpaceX team seems to have chosen another approach, reinforcing the inner ring's power. With five engines, the performance of the steps increases even further. And as the number increases, the proportion of each engine in the total decreases, making the failure of one engine insignificant because other units will compensate for the shortfall. Not only with Super Heavy, but ship may also receive a similar upgrade. We've long seen ship designs with nine engines, when three Raptor vacuums are added. If combined with the current engine configuration, the total Starship engine number could reach 42 engines, which Elon Musk will call Starship V4 in 2025. It will support a longer ship, primarily to extend the fuel tank, and will fly next year. But if the booster also increases the number of engines as indicated, it could reach 44 engines. Overall, it's similar to SpaceX's engine philosophy to date. Choosing to use many small engines instead of a few large engines like traditional rockets. One question is, if a booster version with an increased number of engines can be implemented, what about the engine version? Will SpaceX choose to reinforce the Raptor 3 or create a Raptor 4 instead? The answer is probably that both possibilities are plausible. The continued use of the Raptor 3 is evidenced by another revelation from Elon Musk. Specifically, he stated in May that Starship 5-4 aims to achieve 10,000 tons of thrust, or 22 million pounds of thrust, three times that of the Saturn V. If you divide 10,000 by 35 engines in the upgraded Super Heavy version, it would be over 285 tons, only slightly more than the stated thrust per engine. Therefore, it's still possible that SpaceX will increase the number of engines without changing the version, instead focusing on improving engine reliability. This would be suitable for the near future, as SpaceX continues to launch and hasn't yet released a new engine version. But eventually, a new engine version will certainly have to be announced. This will undoubtedly ensure Starship has over 10,000 tons of thrust, regardless of whether it has 33 or 35 engines. Therefore, Elon Musk stated that each Starship V4 engine will deliver 300 tons of thrust, and you could tentatively call this version Raptor 3X or Raptor 4. Because of this upgrade, the thrust for the V-4 reaches 2,700 tons compared to 1,600 tons in the current V-3. These upgrades give the Starship V-4 a breakthrough in power. Also, in 2025, Elon Musk stated that the V-4, appearing in the next few years, will double the payload of the V-3, specifically over 200 tons compared to over 100 tons. Besides payload, this will also relate to the refueling aspect. Elon Musk stated earlier this year that the Starship V4 tanker could carry over 200 tons of fuel into orbit per flight, significantly reducing the number of refueling flights needed to just five to six flights. Reducing the number of refueling cycles is clearly a major advantage. reducing the rush in preparation, launch, and the number of dockings between ships, as well as addressing the fuel storage issue. That's an improvement in power. In the current context, we expect it to also address reliability. As mentioned, in V-3 flights to date, the engine remains a challenge, preventing flights from reaching perfection. The Raptor 3 is much simpler than the two previous versions. Elon Musk often praises it as engineering art, but he also acknowledges the great internal complexity that only 3D metal printing can create. That's why he advises new companies not to copy the Raptor 3. Not only is manufacturing difficult, but because it's so simple, checking each part and repairing it also becomes challenging. While the method reduces weight and leaks compared to flanges and bolts, it requires cutting out parts for each repair. Clearly, on its path to engine improvement, SpaceX must find ways to address these issues. And right now, the engine upgrade on Starship 5-4 is generating high expectations. That's right. It's not just about increasing the number of engines. It will be about improving reliability to help a flight achieve near perfection. But that's not the ultimate upgrade limit. Don't forget the core and equally important goal, achieving rapid and complete reusability. Because if this problem isn't solved, no matter how powerful the Raptor engine is, launch costs will remain in the tens of millions of dollars, and launch frequency will be stuck at a few months per flight, leaving the dream of building space infrastructure or sending humans to the Moon and Mars on a large scale just a theory. That's why engineers must focus on a completely different approach, designing for full reusability, discarding no parts, and achieving extremely rapid turnaround times of just a few hours to a few days. Look at the Raptor 3 version. It has been thoroughly optimized by eliminating most of the external ductwork and heat shields around the engine, while integrating many components directly into the fuselage to minimize the need for manual inspection or repair after each landing. Combined with the use of a Mechazilla robotic arm to catch the booster stage in mid-air for refueling and immediate relaunch, SpaceX is streamlining rocket operations to a level similar to how a commercial aircraft operates after landing on a runway. The goal of this change is to create a true economic revolution by exponentially reducing launch costs to just a few million dollars, primarily fuel and operating expenses, bringing the cost of transporting each kilogram of material to low-Earth orbit down to a few hundred dollars. Furthermore, a continuous fleet of spacecraft operating at high frequency is crucial for deploying numerous Starlink satellites or performing complex in-orbit refueling missions before undertaking long-duration missions to Mars. And now, as Starship V3 and Raptor 3 face challenges and continue to improve, Starship V4, with its engine changes, may be another step closer to fast and full reusability. Although the V3 generation faced many challenging obstacles regarding stability during takeoff and landing, the overall picture is not pessimistic thanks to the emergence of the next-generation architecture. The current plans and steps are not simply bold statements, but also reflect the thorough strategic vision of the leading engineering team. With groundbreaking advancements in power reinforcement, an increased number of engines and core improvements in durability, reliability, and fuel capacity, the next version is expected to completely eliminate existing bottlenecks. In particular, it is expected to address the super-heavy problem that SpaceX is currently facing. This transformation not only optimizes thrust distribution and enhances safety redundancy for the giant machine, but also provides a superior payload surplus, completely solving the complex in-orbit refueling problem. Thus, the dream of rapid and comprehensive reuse to lower the cost of space transport is getting closer to a realistic goal. If you're not familiar with these satellites, you can basically think of V3 as a massive upgrade over the Starlink satellites being used today. Each V3 satellite can handle up to 1 terabit per second of download capacity, roughly 10 times more than V2, while its upload capacity jumps by around 22 times. It also comes with much more powerful antennas and processing hardware, allowing each satellite to handle far more users and move much more data at the same time. And with upgraded space lasers connecting the satellites together, the entire network can move huge amounts of data across the world. V-3 also uses much larger solar arrays, generating roughly twice the power of the V2 Mini satellites. And all that extra power allows SpaceX to pack in much more powerful communications hardware. But one of the biggest reasons this deployment matters is that V-3 is basically designed around Starship. Each satellite weighs roughly 2,000 kilograms, or about 4,400 pounds, more than three times the weight of a typical V2 Mini. They're also physically much larger, with a bigger satellite bus and much larger solar arrays, making them impractical to launch in useful numbers on Falcon 9. So V3 isn't just a faster Starlink satellite, it's basically Starlink scaling up to a whole new level. More capacity, more users, and a lot more room for the network to grow. Now, all of those upgrades sound really cool for customers like you and me. But what does the U.S. military actually want from all these Starlink satellites? Well, SpaceX now has thousands of Starlink satellites flying around Earth, and that creates a problem. There are now so many satellites and pieces of hardware up there that keeping track of everything is becoming increasingly difficult. Traditional tracking systems might only observe an object a few times a day, And if that object suddenly changes its orbit, it can become much harder to predict exactly where it's going to be next. That's why every Starlink satellite already has star trackers, basically cameras that look at the stars to figure out exactly how the satellite is oriented. The interesting part is SpaceX realized those cameras could also be used to spot other objects moving through space. So the company turned something it already had into a new space tracking system called Stargaze. Instead of relying on a relatively small number of tracking stations, StarGaze can use data from nearly 30,000 star tracking sensors across the Starlink constellation, generating around 30 million object observations every day. That gives SpaceX a much more frequent picture of what's happening around its satellites. The system can then use all those observations to estimate an object's trajectory and quickly calculate whether two objects could come dangerously close to each other. SpaceX says Stargaze can generate collision risk warnings in minutes instead of hours, potentially giving operators much more time to react. Now, there is a catch. Starlink's cameras aren't great at detecting very small objects, which is exactly why high-quality radar and telescope data is still valuable. Ground-based military and commercial tracking systems can routinely track objects down to around 10 centimeters in low-Earth orbit. Stargaze doesn't necessarily match them when it comes to seeing the smallest objects, but that's not really where its advantage is. The big advantage is the sheer number and frequency of observations. Stargaze can take tens to hundreds of pictures of the same object over a period of time, giving SpaceX much more data to work with when figuring out how that object is moving. Those observations are then added to a database, cataloging what objects have been seen, where they are, and how they're moving through space. space. And this is where the really interesting deal comes in. The U.S. Space Force's Joint Commercial Operations Cell and SpaceX have a contractual agreement where SpaceX provides stargaze observations to the government, while SpaceX gets access to highly accurate government space tracking data in return. The limitation is that SpaceX's use of that government data is restricted to internal use rather than commercial exploitation. So basically, the deal is pretty simple. SpaceX gives the Space Force something it already has, millions of observations from thousands of satellites, and the Space Force gives SpaceX something it really needs, highly accurate tracking data from the government. This isn't the only example of the U.S. military increasingly relying on SpaceX's commercial space ecosystem. On May 26th, the Space Force officially announced a $2.29 billion agreement with SpaceX for the Space Data Network Backbone. In simple terms, the U.S. Space Force wants to build a huge communications network in orbit that can move military data around the world quickly and securely. Instead of relying on a small number of traditional satellites and ground stations, the idea is to build a large mesh of satellites that can pass data between each other using high-speed optical links. And SpaceX is now playing a major role in building that network, with the goal of delivering a fully operational prototype by the end of 2027. The interesting part is that this isn't simply about giving the military another version of Starlink. The SDN backbone is designed more like the high-speed backbone of the Internet, but in space. It can move huge amounts of data between military satellites, sensors, and other systems with low latency, while the optical links allow satellites to communicate directly with each other. And because the network is made up of many interconnected satellites, data can potentially be routed around a damaged, jammed, or unavailable part of the network instead of bringing the entire system down. The SDN backbone will also work alongside the Space Development Agency's transport layer, with both systems becoming part of a much larger mesh network in low-Earth orbit. So essentially, the Space Force is trying to build something like a global data highway in space, and SpaceX is being brought in to build a major part of that highway. Then there's also Starshield, SpaceX's government-focused version of Starlink technology, designed for things like secure communications, hosted payloads, Earth observation, and classified processing. It already serves government customers including the Space Force and the National Reconnaissance Office and future Starlink V-3 class satellites could potentially play a role in expanding that network. This is where Starship becomes really important. Higher-capacity V-3 satellites are one thing, but you also need a way to get a huge number of them into orbit, and that's exactly what Starship is supposed to provide. Flight 14 only deployed 26 V-3 satellites, but Starship is designed to eventually carry far more per launch, potentially around 60 V-3 satellites in a single flight. At roughly 2,000 kilograms each, these are massive satellites compared with the older V-2 Minis. But Starship is being designed specifically to handle that kind of payload. If SpaceX can get Starship flying frequently and reliably, it could dramatically increase how quickly these large satellite constellations can be built. So the flight wasn't just about putting 26 Starlink satellites into orbit. It was another demonstration that Starship could become the vehicle that allows SpaceX to deploy these much larger, much more capable satellites at a scale that simply wasn't practical with Falcon 9. And for both commercial Starlink and future government networks, that ability to put a huge amount of satellite capacity into orbit over and over again could be just as important as the satellites themselves. You know what's kind of crazy? The military's relationship with SpaceX isn't just about buying its products anymore, Because now, Elon Musk himself has been brought in to help the Pentagon think about what warfare could actually look like in the future. The story is that on September 30th, Defense Secretary Pete Hegseth announced a new 120-day project called Project Meridian, which is basically the Pentagon asking a pretty huge question. What will the battlefield look like 10, 20, or even 30 years from now? from now? And what kind of technology does the U.S. need to prepare for it? And the Pentagon has brought in some pretty unusual people to help answer that question, with Elon Musk co-leading the project alongside Palmer Luckey, the founder of Anduril, and former House Speaker Newt Gingrich. So, this is kind of Musk coming back into the government world, but in a very different role this time. You might remember that he previously had a much bigger and far more controversial government role through Doge, where he helped lead the Trump administration's effort to cut federal spending and restructure parts of the government. But this time, the Pentagon wants Musk in the room for a very different question. What does the next generation of warfare actually look like? And the list of technologies they're looking at is pretty wild, including AI, autonomous systems, robotics, directed energy weapons, biotechnology, new ways of fighting in space, warfare underground, and even potential military operations around the moon. Basically, the Pentagon isn't just thinking about the next fighter jet or the next missile anymore. It's asking what happens when AI, robots, autonomous systems, space technology, and all these other emerging technologies start completely changing how wars are fought. As we talked about earlier, SpaceX already operates one of the world's largest satellite networks, launches military payloads, develops reusable rockets, and is working on technologies that could eventually support missions far beyond Earth. And Musk isn't the only tech executive involved here either, because Palmer Luckey's Anduril is already heavily focused on autonomous military systems and other defense technologies, while SpaceX already has major government and national security contracts. So now, you have some of the people actually building these technologies Sitting in the room with the Pentagon and being asked to think about what the battlefield of the future could look like. And honestly, that's what makes Project Meridian so interesting. Because the US military isn't just looking at the technology it has today, it's trying to figure out what technology it might need decades from now. And companies like SpaceX and Anduril are becoming a pretty big part of that conversation. And if you want more cool, up-to-date stories like this, hit that subscribe button so you You can be one of the first to know.