My piece on Orbital’s, SpaceX’s and ILS’s trifecta is up now at PJMedia.
10 thoughts on “Super Sunday For Commercial Spaceflight”
For once I find myself agreeing with Jim (about the Senate Democrats OK with a “clean CR” that keeps the Sequester).
The President is slouching towards a self-brought-upon meltdown. His Secretary of State, smart man of history that he is, invokes the spectre of Chamberlain at Munich and then, guess what, gets off a plane waving a piece of paper proclaiming peace in our time based on a dodgy promise from a tyrant. You know the Exchanges won’t work from an IT perspective (a few bugs that can be worked out, remind me Jim about your understanding of Missile Defense software) unless they “pull out all of the safeties” on any kind of verification. Gun Control anyone? Immigration Reform (OK, that one is not quite dead yet).
Just give the President a clean Continuing Resolution — there is no Deal, there is no Grand Bargain, and yes, the optimal strategy is indeed to kick-the-can-down-the-road and muddle through. But noooooooo! First our die hard freedom-loving TEA Party/Libertarian Patriots scared off the working classes (Thank you Governors Walker, Kasich, and Scott along with former Governor Romney blabbing about the unjust Auto Bailout . . . in Ohio! Thank you for helping reelect Mr. Obama as President. And y’all, tell me I am wrong and that what the Right/TEA/Conservative Blogosphere is for, yeah, yeah, double-digit trillion debt, and even Obamacare are the existential threats of our day and no one was scared and only the stoopid voted for Mr. Obama, but last I checked Mr. Obama is still President).
Yes, Mr. Obama is President, and headed nowhere fast, with rot setting in to his ill-begotten empire, fall-of-Michael-Corleone style. But we have to call attention to ourselves, don’t we, we have to give Mr. Obama his renewed Moment in the Sun, we can’t just let whatever he is about collapse?
I must have missed the connection Paul.. is your rant about using Landing Legs on the Falcon 9?
If SpaceX pulls off a landing it will really make the porkonauts in congress look bad. There will no justifing the porkwagon approach anymore.
It will cause an “Oh $hit!” reaction in China, Russia, Europe and Chicago (I almost said Seattle).
And as to the promise of ’69; Elon will build it, but will they come?
Thanks, interesting report. Upgraded engine, stretching the tanks, experimenting with moving the engines into a different configuration: seems to me that this is the sort of attitude NASA should have had with the Shuttle. Now SpaceX is having to do the work for them.
I’m surprised by the lack of reaction to the PJM post. I’ll repeat the comment I made there here:
The more I think about it, the more I believe that they actually came close to achieving a soft landing on the first try, which is absolutely incredible.
I figured that the hardest parts would be the turnaround maneuver and first burn, but they seem to have gone off smoothly. I was half expecting the vehicle to break up at that point. They have already done the landing part with Grasshopper.
Compared to all that, fixing the spin problem shouldn’t be too difficult. Elon said that they now have all the pieces of the puzzle needed to recover the first stage.
I’m surprised by the lack of reaction to the PJM post.
Well, it’s been a busy news day…
September 28, 2008 was the first successful orbital flight of Falcon 1, after three failures.
A lot has happened in five years and one day.
The best part about SpaceX is simply the fact that they have their head screwed on straight. In the aerospace industry there is this pernicious idea that it makes sense to skip generations. You see that with the Shuttle, VentureStar, JWST, the F-35, rotary rocket, Orion, SLS, and so much more. It’s easier to sell an N+2 design because it’s sexier, and because N+2 designs are obviously superior, and because every time it’s claimed that skipping a generation will be faster and cheaper. Yet it never works out in practice. STS was the most expensive heavily utilized launcher in history. And there were so many development problems with VentureStar that it had to be scrapped. I don’t remember it off-hand but they’ve made up a new term for the F-35 to describe the fact that the first models are actually really prototypes.
It’s almost always the same. Skipping generations is slower, costlier, and riskier.
SpaceX is smart enough to realize that there’s no shortcut for the necessary hard work of proving each iteration of a design. Because there are valuable lessons to be learned in each phase and because you mitigate risk by taking smaller steps. If SpaceX were like NASA or most companies they would have put forward their N+2 or N+X design as a multi-core heavy lift booster with every stage reusable and with a deep-space manned spacecraft as a payload. But if they’d tried that they would have spent more time and vastly more money before getting to where they are now and they would have a much lower chance of success. Instead they started out with the Falcon 1, then moved ahead at a development pace that seemed appropriate, evolving the Falcon 9, then the Falcon 9 v1.1, and working away on all the necessary technological advances they need for each phase of iterative improvement. Instead of trying to build a manned spacecraft in one go they started off by testing the key elements of the vehicle design in an unmanned cargo vehicle. Instead of trying to build full stack reusability from day 1 they worked on how to evolve the vehicle toward reusability. Instead of trying to build a heavy lift version as the main offering, they concentrated on proving the core components first.
Even now if you look at what SpaceX has to offer you can see a lot of places where things are sub-optimal. The 2nd stage of their rocket is LOX/Kero so they don’t have very impressive GEO performance. 2nd stage restarts are still a bit buggy. The Falcon 9-R will only reuse the first stage. The manned Dragon is still not finished. And so on. But despite all that they still have a very good rocket and a good pressurized spacecraft that are solid competitors in the market, and their development of stage reusability will make them even more so by a tremendous margin.
Within the next few years SpaceX will have a heavy lift rocket, a reusable first stage at crazy prices, a manned spacecraft, and perhaps a LOX/Methane upper stage. All of which will further accelerate the ability of the company to generate revenue, which feeds into their R&D budget and allows them to improve what they’ve got yet again.
On the one hand it’s quite impressive. On the other hand it seems as though someone could have been developing these sorts of things back in the ’70s or at least ’80s (DC-X comes to mind) but nobody had the combination of authority and pragmatism to make it happen.
I hate replying to myself but I realize there’s an important point I left out of the above.
When developing novel technology, iteration has always been the best and surest route to success. Look at aviation, the automobile, computers, anything. The first automobile was very primitive and even downright problematic compared to modern versions, but bit by bit, component by component, advancement by advancement the designs improved. Look at all of the parts of a modern automobile that weren’t there at the start, an electrical system with a battery that can be recharged by the engine, electric start, head lights, a fully enclosed cabin, anti-lock brakes, seat belts, EFI, windshield wipers, and so on.
And this is why commercial development and a service relationship between government and private enterprise are vastly better models than large government driven (work for hire) development projects. The very nature of a procurement contract imposes a “design ONE thing then build it” restriction on the project. These sorts of projects are fine when building something that’s well known but they are horrible when trying to push the state of the art. The very idea presupposes that the designers know all the answers already without actually having any hands on experience in the problem space. But it’s difficult to pitch the idea of a substantial amount of funding being set aside for a development program which may end up producing something worthwhile or may not.
In the end it’s easier to present a degree of certainty in the ability to make a speculative design work, even when that certainty has no basis in fact, than to attempt to sell a plan that’s acknowledged to be risky. But again, look at the Shuttle or the F-35. Those programs were bought hook, line, and sinker by congress on promises that were magnificently unrealistic and which the programs have missed by orders of magnitude to achieve. For the F-35, for example, the notion was that it would be cheaper and easier to fly a “common” aircraft across different service branches and different governments. In practice the 3 versions of the F-35 only share about 30% of components and really only share a name and a body line, and that’s not even getting into the cost or performance issues.
Compare the Shuttle vs Falcon 9 development, for example. The Shuttle was designed to be a reusable launcher, after hashing out all the requirements and engineering constraints they came up with a design, built it, and then flew largely the same design for 30 years. That process produced a hugely compromised vehicle which ended up being more expensive per launch than the Saturn V while putting a tiny fraction of its payload into orbit. SpaceX took a different tack, they started building hardware and flying rockets only 4 years after the company was founded. The current design of the Falcon 9-R is very different from the early Falcon 5 concepts and even different from the reusable Falcon 9 concepts from only 2 years ago.
It’s very important to highlight what SpaceX did with its most recent flight. It used an operational flight of a rocket as part of an R&D program for a future product. The design of the Falcon 9 v1.1 is driven as much by the need to prove out concepts for the Falcon 9-R as it is to merely make a successful orbital launcher. From the very first second of the launch of the new Falcon 9, with that green flash from the initiators bringing the engines to life, we see that this isn’t just an operational vehicle, it’s also a test vehicle. What reason does an ordinary Falcon 9 v1.1 launch have to restart the first stage engines? But by building in those features with a low overhead and low risk to the nominal mission they were able to test all of these things that will heavily inform the future design of the next iteration of their launcher. By sharing substantial design elements between the Dragon cargo and the manned Dragon they can get a big head start on developing the manned version. They don’t need to prove that the pressure hull of the manned Dragon is reliable, they don’t have to prove the heat shield works, or the integration with Falcon, these things have already been demonstrated operationally. By the time SpaceX looks toward reusability on the 2nd stage it’s quite possible that it’ll end up being quite different from any idea they had at the outset. Perhaps they’ll have switched to a LOX/LCH4 stage by then, perhaps the landing leg design will be informed by experiences and improvements with the first stage legs.
It turns out that when you build and fly hardware you learn lessons and acquire expertise, sometimes very important lessons. Moreover, sometimes these lessons take you to different designs and different tradeoffs than you could have possibly come up with otherwise. In the end the Falcon 9-R is nothing like the original design of a reusable vehicle that SpaceX had 10 years ago, but it’s far more likely be able to achieve the original desired goals in terms of reliability, capability, and cost than any government funded rocket has managed or is likely to manage (Shuttle, X-33/VentureStar, SLS, even EELV).
The best way to move forward in spaceflight innovation is to recognize the mistakes of the past and stop lying to ourselves. Stop believing the promises of these aerospace megaprojects. They have as much chance of being anywhere close to accurate as winning the lottery. Yet we continue to play the lottery again and again and we continue to believe these claims as though we are children who haven’t heard the truth about Santa yet. The only way to innovate is to embrace the reality of risk, if there was no risk it wouldn’t be innovation.
Great comments, Robin.
I’ve said before that I’m more excited by what SpaceX is doing now than I have been at any time since the Apollo era.
After my initial excitement about the Shuttle wore off, I was less than excited about it because a) NASA didn’t have any real plans to go back to the Moon, and b) it was apparent that the Shuttle system wasn’t readily upgradable. It wasn’t like they could stick on a couple more SRBs and fly it to the Moon or Mars.
For once I find myself agreeing with Jim (about the Senate Democrats OK with a “clean CR” that keeps the Sequester).
The President is slouching towards a self-brought-upon meltdown. His Secretary of State, smart man of history that he is, invokes the spectre of Chamberlain at Munich and then, guess what, gets off a plane waving a piece of paper proclaiming peace in our time based on a dodgy promise from a tyrant. You know the Exchanges won’t work from an IT perspective (a few bugs that can be worked out, remind me Jim about your understanding of Missile Defense software) unless they “pull out all of the safeties” on any kind of verification. Gun Control anyone? Immigration Reform (OK, that one is not quite dead yet).
Just give the President a clean Continuing Resolution — there is no Deal, there is no Grand Bargain, and yes, the optimal strategy is indeed to kick-the-can-down-the-road and muddle through. But noooooooo! First our die hard freedom-loving TEA Party/Libertarian Patriots scared off the working classes (Thank you Governors Walker, Kasich, and Scott along with former Governor Romney blabbing about the unjust Auto Bailout . . . in Ohio! Thank you for helping reelect Mr. Obama as President. And y’all, tell me I am wrong and that what the Right/TEA/Conservative Blogosphere is for, yeah, yeah, double-digit trillion debt, and even Obamacare are the existential threats of our day and no one was scared and only the stoopid voted for Mr. Obama, but last I checked Mr. Obama is still President).
Yes, Mr. Obama is President, and headed nowhere fast, with rot setting in to his ill-begotten empire, fall-of-Michael-Corleone style. But we have to call attention to ourselves, don’t we, we have to give Mr. Obama his renewed Moment in the Sun, we can’t just let whatever he is about collapse?
I must have missed the connection Paul.. is your rant about using Landing Legs on the Falcon 9?
If SpaceX pulls off a landing it will really make the porkonauts in congress look bad. There will no justifing the porkwagon approach anymore.
It will cause an “Oh $hit!” reaction in China, Russia, Europe and Chicago (I almost said Seattle).
And as to the promise of ’69; Elon will build it, but will they come?
Thanks, interesting report. Upgraded engine, stretching the tanks, experimenting with moving the engines into a different configuration: seems to me that this is the sort of attitude NASA should have had with the Shuttle. Now SpaceX is having to do the work for them.
I’m surprised by the lack of reaction to the PJM post. I’ll repeat the comment I made there here:
Compared to all that, fixing the spin problem shouldn’t be too difficult. Elon said that they now have all the pieces of the puzzle needed to recover the first stage.
I’m surprised by the lack of reaction to the PJM post.
Well, it’s been a busy news day…
September 28, 2008 was the first successful orbital flight of Falcon 1, after three failures.
A lot has happened in five years and one day.
The best part about SpaceX is simply the fact that they have their head screwed on straight. In the aerospace industry there is this pernicious idea that it makes sense to skip generations. You see that with the Shuttle, VentureStar, JWST, the F-35, rotary rocket, Orion, SLS, and so much more. It’s easier to sell an N+2 design because it’s sexier, and because N+2 designs are obviously superior, and because every time it’s claimed that skipping a generation will be faster and cheaper. Yet it never works out in practice. STS was the most expensive heavily utilized launcher in history. And there were so many development problems with VentureStar that it had to be scrapped. I don’t remember it off-hand but they’ve made up a new term for the F-35 to describe the fact that the first models are actually really prototypes.
It’s almost always the same. Skipping generations is slower, costlier, and riskier.
SpaceX is smart enough to realize that there’s no shortcut for the necessary hard work of proving each iteration of a design. Because there are valuable lessons to be learned in each phase and because you mitigate risk by taking smaller steps. If SpaceX were like NASA or most companies they would have put forward their N+2 or N+X design as a multi-core heavy lift booster with every stage reusable and with a deep-space manned spacecraft as a payload. But if they’d tried that they would have spent more time and vastly more money before getting to where they are now and they would have a much lower chance of success. Instead they started out with the Falcon 1, then moved ahead at a development pace that seemed appropriate, evolving the Falcon 9, then the Falcon 9 v1.1, and working away on all the necessary technological advances they need for each phase of iterative improvement. Instead of trying to build a manned spacecraft in one go they started off by testing the key elements of the vehicle design in an unmanned cargo vehicle. Instead of trying to build full stack reusability from day 1 they worked on how to evolve the vehicle toward reusability. Instead of trying to build a heavy lift version as the main offering, they concentrated on proving the core components first.
Even now if you look at what SpaceX has to offer you can see a lot of places where things are sub-optimal. The 2nd stage of their rocket is LOX/Kero so they don’t have very impressive GEO performance. 2nd stage restarts are still a bit buggy. The Falcon 9-R will only reuse the first stage. The manned Dragon is still not finished. And so on. But despite all that they still have a very good rocket and a good pressurized spacecraft that are solid competitors in the market, and their development of stage reusability will make them even more so by a tremendous margin.
Within the next few years SpaceX will have a heavy lift rocket, a reusable first stage at crazy prices, a manned spacecraft, and perhaps a LOX/Methane upper stage. All of which will further accelerate the ability of the company to generate revenue, which feeds into their R&D budget and allows them to improve what they’ve got yet again.
On the one hand it’s quite impressive. On the other hand it seems as though someone could have been developing these sorts of things back in the ’70s or at least ’80s (DC-X comes to mind) but nobody had the combination of authority and pragmatism to make it happen.
I hate replying to myself but I realize there’s an important point I left out of the above.
When developing novel technology, iteration has always been the best and surest route to success. Look at aviation, the automobile, computers, anything. The first automobile was very primitive and even downright problematic compared to modern versions, but bit by bit, component by component, advancement by advancement the designs improved. Look at all of the parts of a modern automobile that weren’t there at the start, an electrical system with a battery that can be recharged by the engine, electric start, head lights, a fully enclosed cabin, anti-lock brakes, seat belts, EFI, windshield wipers, and so on.
And this is why commercial development and a service relationship between government and private enterprise are vastly better models than large government driven (work for hire) development projects. The very nature of a procurement contract imposes a “design ONE thing then build it” restriction on the project. These sorts of projects are fine when building something that’s well known but they are horrible when trying to push the state of the art. The very idea presupposes that the designers know all the answers already without actually having any hands on experience in the problem space. But it’s difficult to pitch the idea of a substantial amount of funding being set aside for a development program which may end up producing something worthwhile or may not.
In the end it’s easier to present a degree of certainty in the ability to make a speculative design work, even when that certainty has no basis in fact, than to attempt to sell a plan that’s acknowledged to be risky. But again, look at the Shuttle or the F-35. Those programs were bought hook, line, and sinker by congress on promises that were magnificently unrealistic and which the programs have missed by orders of magnitude to achieve. For the F-35, for example, the notion was that it would be cheaper and easier to fly a “common” aircraft across different service branches and different governments. In practice the 3 versions of the F-35 only share about 30% of components and really only share a name and a body line, and that’s not even getting into the cost or performance issues.
Compare the Shuttle vs Falcon 9 development, for example. The Shuttle was designed to be a reusable launcher, after hashing out all the requirements and engineering constraints they came up with a design, built it, and then flew largely the same design for 30 years. That process produced a hugely compromised vehicle which ended up being more expensive per launch than the Saturn V while putting a tiny fraction of its payload into orbit. SpaceX took a different tack, they started building hardware and flying rockets only 4 years after the company was founded. The current design of the Falcon 9-R is very different from the early Falcon 5 concepts and even different from the reusable Falcon 9 concepts from only 2 years ago.
It’s very important to highlight what SpaceX did with its most recent flight. It used an operational flight of a rocket as part of an R&D program for a future product. The design of the Falcon 9 v1.1 is driven as much by the need to prove out concepts for the Falcon 9-R as it is to merely make a successful orbital launcher. From the very first second of the launch of the new Falcon 9, with that green flash from the initiators bringing the engines to life, we see that this isn’t just an operational vehicle, it’s also a test vehicle. What reason does an ordinary Falcon 9 v1.1 launch have to restart the first stage engines? But by building in those features with a low overhead and low risk to the nominal mission they were able to test all of these things that will heavily inform the future design of the next iteration of their launcher. By sharing substantial design elements between the Dragon cargo and the manned Dragon they can get a big head start on developing the manned version. They don’t need to prove that the pressure hull of the manned Dragon is reliable, they don’t have to prove the heat shield works, or the integration with Falcon, these things have already been demonstrated operationally. By the time SpaceX looks toward reusability on the 2nd stage it’s quite possible that it’ll end up being quite different from any idea they had at the outset. Perhaps they’ll have switched to a LOX/LCH4 stage by then, perhaps the landing leg design will be informed by experiences and improvements with the first stage legs.
It turns out that when you build and fly hardware you learn lessons and acquire expertise, sometimes very important lessons. Moreover, sometimes these lessons take you to different designs and different tradeoffs than you could have possibly come up with otherwise. In the end the Falcon 9-R is nothing like the original design of a reusable vehicle that SpaceX had 10 years ago, but it’s far more likely be able to achieve the original desired goals in terms of reliability, capability, and cost than any government funded rocket has managed or is likely to manage (Shuttle, X-33/VentureStar, SLS, even EELV).
The best way to move forward in spaceflight innovation is to recognize the mistakes of the past and stop lying to ourselves. Stop believing the promises of these aerospace megaprojects. They have as much chance of being anywhere close to accurate as winning the lottery. Yet we continue to play the lottery again and again and we continue to believe these claims as though we are children who haven’t heard the truth about Santa yet. The only way to innovate is to embrace the reality of risk, if there was no risk it wouldn’t be innovation.
Great comments, Robin.
I’ve said before that I’m more excited by what SpaceX is doing now than I have been at any time since the Apollo era.
After my initial excitement about the Shuttle wore off, I was less than excited about it because a) NASA didn’t have any real plans to go back to the Moon, and b) it was apparent that the Shuttle system wasn’t readily upgradable. It wasn’t like they could stick on a couple more SRBs and fly it to the Moon or Mars.