At Least It Didn’t Blow Up On The Pad



If you could somehow tune your eyes to see the electromagnetic spectrum while watching yesterday’s spectacular launch of SpaceX’s Super Heavy/Starship, you’d see fat ropes of 0’s and 1’s streaming back to earth. And therein lies the explanation of why the thing blew up four minutes into the flight and the solution to fixing it so there won’t be a repeat. Or at least one that blows up at a different point.

The launch was deemed a success by SpaceX and if so, it wasn’t an unqualified one. Elon Musk lead turned the outcome earlier in the week. “I’d like to set expectations low,” he said. “If we get far enough away from the launchpad before something goes wrong, I think I would consider that a success. Just don’t blow up the launchpad.”

Musk has had a pattern of sorts in saying his rockets are likely to blow up. I’ve always wondered if this is honestly reflective of a culture that moves at the speed of heat with high risk tolerance and minimal testing or just spin against the inevitable failures. It’s probably a little of each. Whatever it is, the SpaceX culture has produced reliable boosters and spacecraft that, if they haven’t revolutionized access to space, have certainly shaken up the status quo. No other company, country or even NASA is close to that kind of achievement recently.

Last November, we watched NASA finally launch its long delayed Space Launch System booster with an unmanned Artemis topping the stack. Its contrast with the Starship system couldn’t be starker. The SLS emerged from NASA’s typical plodding, circumspect and sometimes risk-averse development ethos. It was supposed to fly in 2016, but launched six years after that. SpaceX has famously fictional launch predictions so it goes with the territory, I guess. On the other hand, composed as it was from repurposed Space Shuttle parts, the SLS launched flawlessly and yielded a highly successful if not visionary mission.

I was watching Miles O’Brien on PBS last night who seemed a little puzzled at why all those SpaceX people were cheering and cheered even louder the booster went awry and blew up. I had the same reaction. If you’ve poured blood and sweat into a project this momentous, wouldn’t you be on fingernail-chewing pins and needles watching it fly? I would. But then I figured it out. At least I think I did. They were just thrilled to have gotten the thing to the point where it could sit on the pad awaiting ignition. As the SpaceX narrator said, anything past clearing the tower is icing on the cake. To those of us steeped in 1960’s space development, that’s like saying, well, at least Neil got it on the Moon before he hit a boulder and dumped the thing over, stranding two astronauts for eternity. We were primed and expecting of success and would have been devastated by failure. All of us a certain age remember the pain of watching our early rockets blow up. Maybe the SpaceX people expected a pad explosion and would have been unperturbed by it; anything better than that was a thrill.

Now they have to figure out how to make 33 engines start, run and keep running until the Starship has sufficient altitude and velocity to reach orbit with its own engines. SpaceX today reported “multiple” engine failures, and photos of the exhaust end showed at least five weren’t burning. That’s 15 percent loss of thrust and caused the vehicle to go off trajectory and lose altitude. (The engines that were running, weren’t at full power.) Presumably, all that data will yield some answers on the how and why of engine failures. But no one should underestimate the difficulty in keeping that many engines burning in close proximity at once. Just the fuel plumbing alone is daunting, not to mention the shock waves and resonance playing hell with combustion stability.

As they raced the U.S. for the Moon, the Soviets tried the multiple engine approach in the N1 rocket, the counterpart to the American Saturn 5. The N1 had 30 engines and never flew successfully, with one explosion causing a crater later visible by satellite reconnaissance. The engines, it turned out, had multiple interactions that were never sorted out because the Soviets didn’t static fire the system in its flight configuration. SpaceX did static fire the Super Heavy, but only for 10 seconds. Maybe Thursday’s flight produced enough data to not require further static fires. Computer modeling can do wonders with even minimal data.

SpaceX may also need to think about a flame trench and deluge system to tamp down resonances, as NASA did with the Saturn launches. Musk said his biggest worry about a pad explosion was that it would obliterate SpaceX’s ground facilities along the beach at Boca Chica, Texas. News reports say the facility suffered significant damage. It’s also possible that debris from the crater the exhaust dug may be implicated in control or engine failures. While SpaceX is tidying up that mess, they’ve got more Super Heavys on the production line ready for another try.

Other AVwebflash Articles


  1. Overlooked is the fact that this rocket was built out of welded stainless steel, outdoors, in a field in Texas. Compared to the old way of building rockets, doing it this way means that the decrease in cost to orbit with Starship will be massive.

    The systems complexity issues are not to be underestimated, but the comically massive profit potential of Starlink means that they’ll have effectively unlimited money to get them figured out.

    Reusability will be icing on the cake. That part is still less clear to me. They plan to use heat shield tiles which should work fine for low earth orbit, but using them over and over again, I don’t know. And using tiles for lunar and interplanetary reentry seems like it’s a bridge too far. But without reusability, they could make the engines quite a bit cheaper, and that will help even more.

    It’s going to work. And it’s going to change space forever.

  2. It did seem a little odd that everyone was celebrating and doing high fives following the destruction of the booster and starship. But one encouraging thing that did show up from that last-minute video was that the whole assembly held together even after tumbling out of control. I realize that SpaceX wanted to move quickly with the booster by using 33 of the reliable Raptor engines rather than developing an expensive, larger engine, but the plumbing and sonic resonance of all those engines in close proximity can be a huge problem. Having said that, I’m putting my money on the theory that flying debris from the launchpad may have struck and damaged the booster before it even lifted off. If the on-screen videos were accurate, at least three engines failed almost immediately after liftoff and you could see jets of flame popping out away from the booster very quickly – sure signs that something was amiss. Damage photos of the pad show major structural damage and a huge crater underneath the support structure. All that concrete had to go somewhere. SpaceX may have another booster and Starship ready in a couple months, but getting the pad repaired and building a proper blast trench might take a lot longer.

    • Yes, pad repairs may take some time, though a little less than some suggest, given that upgrades are already partially built and were waiting for this launch before they were to be installed.
      Presumably we won’t see another pad failure like this due to lessons learned. And more boosters and ships are stacking up (excuse the pun) waiting for their turn to fly, so cadence could get fast once these initial issues are sorted.

  3. It is fun until someone is killed, or hurt enough for lawyers to have payday. People are a lot less forgiving about flying things falling on them, than they are even of “auto-pilot” cars crashing into things.
    The way Space X moved on to land and forced neighbours away raised eyebrows, even in Texas — raised eyebrows in Texas can quickly escalate into court room drama.

  4. I’ve always wondered why any unmanned rocket isn’t launched with the hypersonic (magnetic drive) method to get them out of reach of earth’s atmosphere. They wouldn’t have to be launched vertically. The savings of weight, cost and safety would go up greatly. Bigger is truly not always better!

  5. My guess is that the purpose of this launch was data capture for systems validation. The engineers were ecstatic because of the all the data they collected. A four minute fight provided terabytes of systems data. Running 33 rocket engines at the same time is a massive systems engineering problem there and there is no way to validate on the ground. Testing one to five engines, the engineers can understand how the engines will perform individually and the interactions between them which comprises the system. When 33 engines are clustered, understanding the interactions is a massive problem. Each engine is producing vibration, stress, noise, heat, thrust vectors, etc. All of these forces are interacting with each other and the effects are not linear. Ever wonder why NASA added solid rocket boosters to the space shuttle instead of scaling up the liquid fueled engines? This launch demonstrated the answer.

  6. The “top down” method instead of “bottom up” is typical Musk common sense. Assuming no humans are involved, and given the available (private) funds, why test something that is half of what you need? Test what you need, find out how far away from success you are, fix that, and try again. My experience, (having extensive experience in racing engines and dynamometer testing, no humans), this method was the shortest way to our team finding maximum success. Remember it’s “not your money” (public government funds). His goal, his money, and his risk. Even the launch pad is now stress tested and will be rebuilt to the “spec” that will be needed for success. Brilliant.

  7. Paul, my sense is that the enthusiasm of the SpaceX team, despite all the failures (five engines out, launch pad destroyed, rocket loss of control and self destruct) is a reflection of SpaceX’s extraordinary culture that embraces failure as a necessary and powerful part of the learning and innovation process. I think there is a bigger story just on that for you to consider. The fact that Falcon 9 has reduced the cost of access to orbit by an order of magnitude, and done so in a fraction of the time NASA takes to do almost anything.
    Time is Money, apparently, and this culture seems to produce incredibly safe and reliable systems, even if early iterations are fraught with failures. SpaceX flies our astronauts to the ISS, not NASA. It’s the speed of the iterative process that ultimately gets them to great engineering much faster. And, I think that is worth studying and learning from, in all areas of aerospace. I look at the aging GA fleet and 50 year old designs, like new Bonanzas costing $1M and I have to think we need more of SpaceX’s efficiency and first principles methodology if we are going to keep GA alive and affordable. I’d love to hear your perspective on this.

  8. This rocket was a test article. A lot of lessons learned in building it have already been built into later iterations. It was flight ready, but just barely. The choices were to just scrap it, or fly it and get whatever data they could get before something went wrong. The same way they iteratively built and launched Starship upper stages, testing the aerodynamics of a novel re-entry profile. The landing algorithms were not ready, and it showed–spectacularly. Until they got it right.

    One thing is certain. It’s a lot of fun to watch–from a distance.

  9. I was amazed that there are 33(!) rocket motors for propulsion. As a retired A&P just the ungodly amount of fuel connections would be a plumber’s nightmare. Rocketry is still fraught with hazards and Musk and his team will try, try again.

  10. Looked to me that the most serious problem was column buckling as the first stage tank emptied. The buckling took place at the juncture between the first and second stages, staring with a pronounced lateral oscillation followed by a jack knife. Together the two stages are a slender column, the first serving as a fluid column surrounded by a thin steel tube. As the first stage propellant burned off, the tube became incapable of supporting the compression between the thrust of the rockets and the several-g load of the second stage. If that is indeed the cause of the failure, the fix might be throttling back the first stage to prolong the time to separation. Steve Crow

  11. Flew my RV6 from AZ to watch this launch and see StarBase, and one word, IMPRESSIVE. Driving right up to where this massive, silver ship sitting on its pad and seeing the other SpaceX operations going on all around it is really something remarkable Its almost like some advanced alien race has set up shop here on earth and us mere humans get to watch the fun.

    Yes the StarShip Heavy failed before reaching orbit, just like the early Falcon rocket failures that nearly bankrupted Musk, but with very large brass and incredible talent he and SpaceX keep going and now the Falcon 9 and Falcon 9 Heavy are the most cost effective, successful, REUSABLE Launch vehicles ever built, full stop. There are national space programs, including NASA, that haven’t done what SpaceX is doing now routinely.

    As a side note, due to thousands of people in the area there was no cell internet service at the viewing area on South Parde Island, except for those who had a SpaceX StarLink antenna, fed by satellites launched by SpaceX!

    My utmost congratulations and respect go out to Elon Musk and the crew at SpaceX. They will undoubtedly fix the problems with StarShip and cant wait to see their next amazing achievement.

  12. Judging by the apparent size of the hole that the first stage dug under the OLP, looks like they have a head start on an awesome flame diverter pit. I am not alone in thinking, from the beginning, that the lack of this obvious feature was a mistake.

  13. We’re forgetting the spectacular failures of our own NASA booster efforts back into the 50s. Nothing is 100%. Space-X will work it out.

  14. When you’re the richest man in the world, the cost of failure is not even a bump in the road, especially when you have investors willing to share the burden. A launch pad destroyed? Not an issue. Lives are lost? Put it behind you and move on. Even if an entire Mars crew is lost on the launch pad it will not stop progress. It’s just the way we operate as humans. A little fodder for the canon is SOP.

  15. Just a nit, I guess, but I was completely confused by the opening sentence. Light is a part of the EM spectrum, so the “somehow” made no sense to me at all. I had to read it multiple times to figure out what was actually intended.