Musk on Starship, engineering, and management
Several listeners have sent me the link to a fifteen minute interview with Elon Musk that took place the day of his big Starship speech on September 30.
I have embedded it below. Watch it. It reveals several very important overarching things about SpaceX and Musk.
1. Musk calls himself SpaceX’s chief engineer, and during this interview he demonstrates why. He understands this stuff at more fundamental level that I think most similar big rocket company heads. This gives him the ability to distinguish good engineering from bad, and thus shape the company’s design direction more forcefully. It also makes him more similar to the early owners of American airplane companies (Douglas, Boeing, McDonnell, Northrop, etc), all of whom were engineers first and managers second. Too often today CEOs know little about the engineering behind their company, and therefore can be easily sidetracked by bad ideas.
2. At several points in discussing his management approach, Musk clearly wants to give an example of how bad management leads to bad engineering, but it is clear he censors himself. I suspect he is thinking of SLS, but does not want to say so to avoid a controversy he doesn’t need.
3. Much of the interview revolves around the aerospike nozzle, and why SpaceX hasn’t used it. It appears Musk’s reason is that while it might make the exhaust of a rocket more efficient, it causes a loss of efficiency in combustion, and the trade-off isn’t worth it.
4. Finally, Musk’s openness to new ideas, even if they prove him wrong, is quite obvious. As he says,
If someone could show that we’re wrong, that would be great. If someone can show you a way to make your design better, this is a gift. I would be like, thank you! Wow, this is awesome. The worst thing would be that we want to do a dumb design and stick with our dumb design. That would be insane.
This is one of those moments where I think he is thinking of SLS, but doesn’t come right out and say so.
On Christmas Eve 1968 three Americans became the first humans to visit another world. What they did to celebrate was unexpected and profound, and will be remembered throughout all human history. Genesis: the Story of Apollo 8, Robert Zimmerman's classic history of humanity's first journey to another world, tells that story, and it is now available as both an ebook and an audiobook, both with a foreword by Valerie Anders and a new introduction by Robert Zimmerman.
The ebook is available everywhere for $5.99 (before discount) at amazon, or direct from my ebook publisher, ebookit. If you buy it from ebookit you don't support the big tech companies and the author gets a bigger cut much sooner.
The audiobook is also available at all these vendors, and is also free with a 30-day trial membership to Audible.
"Not simply about one mission, [Genesis] is also the history of America's quest for the moon... Zimmerman has done a masterful job of tying disparate events together into a solid account of one of America's greatest human triumphs."--San Antonio Express-News
Several listeners have sent me the link to a fifteen minute interview with Elon Musk that took place the day of his big Starship speech on September 30.
I have embedded it below. Watch it. It reveals several very important overarching things about SpaceX and Musk.
1. Musk calls himself SpaceX’s chief engineer, and during this interview he demonstrates why. He understands this stuff at more fundamental level that I think most similar big rocket company heads. This gives him the ability to distinguish good engineering from bad, and thus shape the company’s design direction more forcefully. It also makes him more similar to the early owners of American airplane companies (Douglas, Boeing, McDonnell, Northrop, etc), all of whom were engineers first and managers second. Too often today CEOs know little about the engineering behind their company, and therefore can be easily sidetracked by bad ideas.
2. At several points in discussing his management approach, Musk clearly wants to give an example of how bad management leads to bad engineering, but it is clear he censors himself. I suspect he is thinking of SLS, but does not want to say so to avoid a controversy he doesn’t need.
3. Much of the interview revolves around the aerospike nozzle, and why SpaceX hasn’t used it. It appears Musk’s reason is that while it might make the exhaust of a rocket more efficient, it causes a loss of efficiency in combustion, and the trade-off isn’t worth it.
4. Finally, Musk’s openness to new ideas, even if they prove him wrong, is quite obvious. As he says,
If someone could show that we’re wrong, that would be great. If someone can show you a way to make your design better, this is a gift. I would be like, thank you! Wow, this is awesome. The worst thing would be that we want to do a dumb design and stick with our dumb design. That would be insane.
This is one of those moments where I think he is thinking of SLS, but doesn’t come right out and say so.
On Christmas Eve 1968 three Americans became the first humans to visit another world. What they did to celebrate was unexpected and profound, and will be remembered throughout all human history. Genesis: the Story of Apollo 8, Robert Zimmerman's classic history of humanity's first journey to another world, tells that story, and it is now available as both an ebook and an audiobook, both with a foreword by Valerie Anders and a new introduction by Robert Zimmerman.
The ebook is available everywhere for $5.99 (before discount) at amazon, or direct from my ebook publisher, ebookit. If you buy it from ebookit you don't support the big tech companies and the author gets a bigger cut much sooner.
The audiobook is also available at all these vendors, and is also free with a 30-day trial membership to Audible.
"Not simply about one mission, [Genesis] is also the history of America's quest for the moon... Zimmerman has done a masterful job of tying disparate events together into a solid account of one of America's greatest human triumphs."--San Antonio Express-News
I vote that the next Starship update should be an interview with Tim. Elon seemed much more relaxed and less ‘stuttery’ here than during the presentation itself.
To me, he sounded as stutter and stream-of-consciousness here as he did in his presentation, once he got past the opening. At the beginning of the presentation, he just seemed overwhelmed by the moment. Also, I think his mind runs way faster than his mouth.
Worker detach Starship Mk1’s upper part again from its lower part again as awaited.
https://www.youtube.com/watch?v=J3dC4pImX1c
From Elon Musk in the interview: “Maybe that reason isn’t valid.”
That is the reason why he thinks differently. Question the assumptions, and the answer “that’s how we have always done it” is a terrible answer; it does not answer the question and shows that the engineer does not know how or why his system is the way that it is.
In my engineering career, the hardest part was finding the right question to ask. Next came finding the right person to answer it.
Musk’s openness to new ideas is obviously why the Starship design has changed so much in the past three years (four different descriptions), although at this point in the development they are committed to quite a few fundamental parts of the design.
The discussion of the aerospike engine is interesting, as it revolves around the combustion efficiency. Ultimately, however, it is the exit velocity of the propellant that matters most for fuel efficiency and payload optimization, but presumably combustion efficiency has an effect on that exit velocity.
An example of combustion inefficiency is the Saturn V’s F1 engine. The flame that is seen coming out of the engine is glowing soot, just like the yellow flame of a candle. The soot is unburned carbon chains from the Kerosene fuel. That carbon had not combined with oxygen inside the engine, so the maximum energy contained in the fuel was not released to provide the theoretical maximum possible velocity of the exhaust.
My understanding of the aerospike engine is that it is best used for single stage to orbit rockets (SSTO). Generally, it is a compromise between a sea level engine and a vacuum engine and better than either engine would be if used on an SSTO from sea level to vacuum.
Feel free to question that assumption, because that is how I have always thought of using it.
Edward: It seems to me that SpaceX’s multi-engine approach, even with a single-stage-to-orbit spacecraft, solves the nozzle issue that the aerospike is trying to solve. You simple put different nozzles on different engines, for use at different times during flight. Musk more or less says this when he talked about using different nozzles on the first and upper stages of rockets to better fit their environments.
Robert,
True, but SpaceX is using two stages rather than one. I have mentioned before that SpaceX is willing to forgo some efficiencies in order to reduce costs. With Starship, they are showing innovation, because they are doing something that does not depend upon their previous hardware. They use Merlin and Merlin Vacuum engines when it makes sense, and they use their new Raptor engines when they make sense.
The downside to having the different engines is that they are dead weight when they are not being used, such as boost back, reentry, and landing burns. The reusability concept makes having the dead weight necessary, since reusability reduces the cost. To not drop unused engines is to forgo a fuel efficiency that expendable rockets have — at greater cost to the customer.
SpaceX developed one basic Starship design that can be used for multiple purposes with only a little modification. These are efficiencies that help to reduce the overall cost of development, and the similarities between manned, cargo, and tanker should reduce the cost of operations, too.
I think that the rocket efficiency is an assumption that engineers used for decades, but the reusability of the rocket is a better efficiency overall that eluded the engineers’ imaginations. Or if they imagined it, they were overruled by their bosses, as when the Space Shuttle did not turn out as inexpensive and re-fly-able as planned resulting in Orion becoming a one-and-done spacecraft, Apollo style.
“It also makes him more similar to the early owners of American airplane companies (Douglas, Boeing, McDonnell, Northrop, etc), all of whom were engineers first and managers second. Too often today CEOs know little about the engineering behind their company, and therefore can be easily sidetracked by bad ideas.”
This is a great point, and so true. That first generation of aerospace chiefs . . . so different from what prevails today.
Of the legacy company CEO’s, I think only Tory Bruno has that kind of engineering savvy, and of course he doesn’t even own the company – worse, he’s handcuffed by his stakeholders, Boeing and LockMart, who see ULA simply as a cash pipeline. Otherwise, it’s something you only see now in the startup owners of the 21st century, led by SpaceX.
It also helps when you don’t have public shareholders to answer to. Going public would handcuff Elon in a lot of ways in these kind of engineering decisions.
Edward:
The aerospike engine type major problems are not, if I remember correctly, a too low combustion efficiency at first place, but that it has a very unfavorable geometry for the effective regenerative cooling of its combustion chamber and nozzle and also builds much heavier than the conventional type of liquid rocket engine, which has a bell-shaped type of nozzle and inner gas expansion.
Elon Musk is not always right in every detail, even if we should admire him. He also cannot know everything and cannot be an expert in every technical field, even if we get this impression often. He has often also quite bad ideas (tunnel boring, hyperloop for example). However, the idea to make the Starship from stainless steel was a good one, but we do not know for sure, if he was originator of this idea.
Thankfully Tim didn’t start screaming and jumping around.
Placing A Starship On A Pedestal
https://www.youtube.com/watch?time_continue=1&v=0U2lcrsUNqM
NASASpaceflight says: “SpaceX’s Starship Mk1 was demated on October 1, 2019 to allow for preparations to take place to prepare for her maiden launch”
space googlevesaire: “Inside Starship cargo bay. Header tanks mounted in tip of nosecone to offset engine weight at rear.”
https://www.youtube.com/watch?v=xP04LGBWmBY
Questioner:
“The aerospike engine type major problems are not, if I remember correctly, a too low combustion efficiency at first place, but that it has a very unfavorable geometry for the effective regenerative cooling of its combustion chamber and nozzle and also builds much heavier than the conventional type of liquid rocket engine, which has a bell-shaped type of nozzle and inner gas expansion.”
You’re falling into the exact trap Musk is making the point about. “Oh, it’s too heavy. It’s hard to cool. But we’re good engineers, give us time and a budget and we can solve those engineering problems.” Musk stops right there: “OK, engineering problem, we can solve those with enough time and money. But SHOULD we? This type of engine does not provide a benefit we need, so NO.”
On the combustion inefficiency, I had not previously heard that, but of course I’m not a rocket engineer that would have actually looked into the documentation. If Musk comes up with a statement like that, especially one that indicates that he did study the materials or talk to engineers that have, I’m going to assume he has a basis for it.
Questioner wrote: “The aerospike engine type major problems are not, if I remember correctly, a too low combustion efficiency at first place”
The chosen fuel is a major factor in combustion efficiency. This was the purpose of mentioning the kerosene fueled F1 engine, in my original comment. I should have gone with my original plan, used more words, and included an example of a hydrogen fueled engine for comparison. This is an example of losing clarity for brevity.
“However, the idea to make the Starship from stainless steel was a good one, but we do not know for sure, if he was originator of this idea.”
My first guess is that it came from someone else during a frank discussion in which the accepted assumptions were questioned. Was a lightweight skin really better than one that can handle high temperatures? This may even have come from a discussion about ways in which to speed up production of the test models: welding aluminum or steel would be faster than winding fibers. Welding outdoors would be faster than waiting for the buildings to be completed. Etc.
Some of my most creative solutions came from wild thoughts on the topic. Using a hibachi indoors is a bad idea, but someone else took that thought and realized we could borrow the hot plate from the breakroom in order to heat the unheated portion on the bottom of a vacuum chamber. As with everyone else’s ideas, most were unusable, in the end, but the gold nuggets are worth the tailings. Throw the bizarre ideas into the group and see if someone can run with it.
The chief engineer does not have to come up with all the ideas. He has to determine which ideas to run with. Often, this requires open discussion. A leader who allows for wild ideas to be presented and considered without shooting them down right away allows for a thinking process that can combine ideas that go against accepted practices. Scaled Composites’ airplanes look like they were designed by thinking differently, too. The X-Prize demanded thinking differently in order to find a better way to design, make, and operate launch vehicles.
SpaceX may be the low cost leader now, but their decisions also leave room for future improvement. These improvements may come from other companies in the next few decades, leaving room for competition to grow. SpaceX’s decisions revolve around low cost operations now rather than low cost operations a decade or two from now.
The Space Shuttle was meant to be a low cost solution a decade from initiation, but it failed even at that. NASA then chose to not improve on what they had and let the disappointment continue for three decades, and now rather than improve they chose to go back to the Apollo model. The throwaway Saturn V and Saturn 1B could have done a better job for less cost than the Shuttle; it is too bad that the Apollo Applications program was scrapped back then, but now it has been made obsolete by Falcon, Dragon, and Starliner.
“Thankfully Tim didn’t start screaming and jumping around.”
Yep. I would have.
Every aerospike I’ve ever seen, physical or graphical, just looked way massive. I know that’s very subjective and naive, but they sure give an impression of adding a large amount of mass for a small increase in overall efficiency.
Starship Mk1 status:
Meanwhile, lower propulsion (tankage + engine section) was placed on a new built mounting jig, in order to ease installation of other flightworthy Raptor engines and potentially also other legs, not to speak about all that missing plumbing.
https://www.youtube.com/watch?v=-aDOpyUmfL4