22 thoughts on “The Dinosaurs Of The Launch Industry”
Question: How much of the re-usability of a F9 is determined by the hardware and how much by the software?
From a hardware standpoint it may be harder for the dinosaurs to catch up but the barriers might be much lower on the software end.
When SpaceX did their first stage test a couple launches back, it was a big deal that they tried to restart the first stage and how this hadn’t been tried before by other companies. But my question is why? Is it because of the limitations of the hardware or is it because the software hadn’t progressed far enough to allow it?
I suspect a major reason why, is that they didn’t need to in order to fulfill contract requirements. The government was going to buy those launches and the companies would make a nice profit regardless so why go the extra steps for no additional compensation? That is the type of thinking that comes out of MBA finance classes as opposed to the more entrepreneurial approach of SpaceX.
I might be off base here someplace but with SpaceX demonstrating that it can be done, other companies might follow suit and find that it is easier to do today than it was twenty or thirty years ago, especially if there are not major limitations in adapting the hardware.
That is the type of thinking that comes out of MBA finance classes as opposed to the more entrepreneurial approach of SpaceX.
That’s not my experience of MBAs. Maintaining a competitive edge is a necessary fact of every business. The failure of some parts of the launch industry to innovate is a result of government interference, nothing more. In the areas of the industry where the government has traditionally had less interest, we’ve seen greater innovation, and in the areas of the industry where the government is the primary customer, we’ve had very little. This has nothing to do with “MBA finance classes”.
Many people inside ULA, for example, have proposed reusability experiments which would cost very little and introduce very little risk to operations. For example, I remember reading about some CORONA-style in-air recovery techniques being proposed for Delta core stage recovery. Why didn’t it happen? Simply: ULA isn’t free to spend their own money. This is a result of the consent agreement that was put in place to allow the merging of the two launch vehicle efforts of Boeing and Lockheed Martin. It was put in place to ensure dissimilar access to space for the US government. Innovation has been systematically stifled. No amount of entrepreneurial spirit at ULA can change this.
I’ve been pondering the possibility of filing a suit with the FTC to force a divestiture of ULA from Boeing/Lockmart.
Overdue. A recent example of the pressure from the parent companies on ULA was “Inspiration Mars” imploding with the “necessity” of using SLS. Though from my sources, Tito was dedicated to SLS, maybe he would have felt _he_ didn’t need it if ULA had been able to provide an alternative in spirited fashion.
Then there’s the despicable story from a year or two back concerning a certain former Congressperson…
Rand, Charles,
I don’t know how likely winning a suit like that would be, but it would be neat to see what ULA could do to compete if it were privately held by someone other than Boeing and LM. They’d still have to deal with a semi-neurotic main customer (NRO), and would need to have the funds to do innovative development. But they definitely have the creativity and talent to give SpaceX a run for its money if they weren’t chained down by two parent companies that often directly compete with them.
~Jon
“That’s not my experience of MBAs. Maintaining a competitive edge is a necessary fact of every business. ”
Not necessarily. Until recently ULA has had a virtual monopoly on government launches. This means there is no pressure or incentive to have a competitive edge. There was hardly even any competition that would determine what a competitive edge is. And because the government is their customer, they wont make changes to their launchers unless the government wants them to. So while that is in part due to the role of government there is also a role played by business philosophy in making these decisions.
Any new capital project would have to meet an IRR and then compete against all of the other projects that also beat the hurdle. A company with a government monopoly has a guaranteed profit and it is stable. It wouldn’t make fiscal sense to be developing re-usable technology because it represents a huge capital investment without any readily apparent returns or more importantly, returns greater than the alternatives.
“Many people inside ULA, for example, have proposed reusability experiments which would cost very little and introduce very little risk to operations.”
And those programs would have to go threw a process that weighs their ability to make profit over other proposed projects and existing projects. I am not familiar with the project you are talking about. It may very well have turned out to be inexpensive but the process to get to implementation sure wouldn’t have been cheap. And why even bother if the customer doesn’t want it or it isn’t part of the current deal?
I bet ULA would be more than happy to build re-usable rockets if the government paid them too.
“Why didn’t it happen? Simply: ULA isn’t free to spend their own money.”
I am also not familiar with the consent agreement that you mention but there is someone who does tell them how to spend their money. And in the case of restricting ULA, it is probably to protect the parent companies.
So, I agree that the government plays a large roll but I also think business does as well. Business people have to operate in the environment created for them, especially if your customer is the government. Sometimes that means maintaining the status quo as a means to preserve steady profits.
This is why I am hopeful for SpaceX. Maybe they can shake things up and get these older aerospace companies taking some risks and building things out of competition instead of merely fulfilling cost plus contract requirements.
I suspect that re-usability such that SpaceX is attempting is really something that needs to be designed in from the beginning. There must be 10,000 things that need to be accommodated in the booster design to even begin to attempt a soft landing — the ability to restart the engine just a small portion of those things.
The thing about the dinosaurs is they have deep pockets and the engineering talent to create competition for a reusable Falcon 9. What’s unknown is whether they have the will.
I liken this to Henry Ford using the assembly line to mass produce automobiles. How much of his contemporary competition simply pulled up stakes and went out of business, and how many learned from observing what Ford was doing and quickly game him competition.
If you plan to land on land, I suspect that’s true. But there were reasonably well-developed plans to try reusing the first stage of the Saturn V, which certainly wasn’t designed for it; the problem was that it made no financial sense unless it was going to make at least a few dozen flights.
Plus, realistically, given it was going to land in the sea, I suspect they’d need a bunch of flights just to figure out how to protect it from the sea water well enough to fly again.
However, I suspect most of us would like to see real competition for SpaceX. We just don’t think the existing launch companies have much chance of doing so.
In fact the last (essentially) fully reusable design in the (downward) evolution of Shuttle from optimal reusable systems use a fly-back ( yes, winged) version of the Saturn V first stage.
When that was rejected because of costs, they went to an external tank/strapon booster configuration, except the strapons were liquids and had clamshell doors designed to protect the motors from seawater. They were intended to be reused, but who knows how practical it would have been.
Finally, when _that_ design was rejected because the peak year development budget (in early ’70’s dollars) was above $1billion, they went to the solids, and convinced themselves that they could be cheaply reusable. And safe.
We all know how that turned out.
“I suspect that re-usability such that SpaceX is attempting is really something that needs to be designed in from the beginning. There must be 10,000 things that need to be accommodated in the booster design to even begin to attempt a soft landing — the ability to restart the engine just a small portion of those things.”
That would speak to how forward looking SpaceX is. Perhaps with SpaceX leading the way, other companies will find it isn’t as hard as they thought it was thirty years ago.
Much of it is dependent on hardware. You need to have an engine that you can restart, which is rarely the case for first stages, you need to have the right engine configuration to be able to provide thrust in the right range to facilitate hovering and also have enough control authority for targeted re-entry and landing, you need the stage to be strong enough to survive entry and descent.
That’s why the F9 v1.1 use TEA-TEB igniters, why it uses the “octo-web” engine configuration, why the central engine is gimballed, and why it has supports and mount points for landing legs. It is simply not the case, not by a long shot, that you could update the software of an Atlas V, Delta IV, Soyuz, or Ariane 5 and enable an RTLS flight profile for the first stage, even aside from the landing gear issues.
The Falcon 9 v1.1 / Falcon 9-R was engineered from the ground up for reusability, the fact that it makes a dandy expendable first stage as well is a double use benefit that proves the genius of SpaceX. They are going to be cranking out v1.1 stages from their factory, paid for by the customers of their expendable launch services. Once they’ve validated the reusable flight-profile of the v1.1 stage they’ll already have hardware on hand, they’ll already have some flights they can reuse the stage from, and they’ll already have their manufacturing fully geared up for production. It’s a shockingly smart business play, and it’s almost certain to disrupt the entire spaceflight industry almost overnight unless some unknown calamity happens to the company.
“That’s why the F9 v1.1 use TEA-TEB igniters, why it uses the “octo-web” engine configuration, why the central engine is gimballed, and why it has supports and mount points for landing legs. It is simply not the case, not by a long shot, that you could update the software of an Atlas V, Delta IV, Soyuz, or Ariane 5 and enable an RTLS flight profile for the first stage, even aside from the landing gear issues.”
Excellent, thanks!
“The Falcon 9 v1.1 / Falcon 9-R was engineered from the ground up for reusability, the fact that it makes a dandy expendable first stage as well is a double use benefit that proves the genius of SpaceX.”
Yes, very impressive and it should go a long way in removing any doubts that they are running the company as some hobby shop out of a dude’s basement.
” It’s a shockingly smart business play, and it’s almost certain to disrupt the entire spaceflight industry almost overnight unless some unknown calamity happens to the company.”
I would add the disruption comes not just from the re-usability but through their pricing strategy. It would have been very easy to undercut their competitors by $10m and rake in the contracts instead of offering drastically lower prices.
The great thing about SpaceX is that they aren’t just a company who builds rockets as an excuse to make money, they’re a company that makes money as an excuse to make rockets. That sort of passion has lots of follow through benefits that puts them well ahead of the game but also gives them market advantage (the way any creator with a legitimate passion for the field they’re in has).
If Musk merely wanted to make a profit of an average of, say, a billion a year over the next decade he could easily do that without reusability, without the MCT, without the manned Dragon. But it’s those things which will propel SpaceX even farther ahead of its competitors over the next several years, and in large part they are built out of a sense of vision and imperative.
For starters having 9 first stage engines means that the throttling requirements are much less. You really only need one throttleable engine. Try that with an Atlas 5.
I know there are other methods than a vertical powered landing.
I didn’t follow this. Could you explain it? – Thanks.
Not sure how many engines Delta has, but I believe the point is that if, say, you need 10% thrust to land, a stage with one engine would have to throttle it down to 10%, while a stage with nine engines would only have to throttle down to 90%.
When you have multiple engines, you can shut the unnecessary ones down.
Almost all of the first stage mass is fuel. Engines capable of >1g acceleration of a fully fueled first stage plus the upper stage(s) plus the payload will have far, far too much thrust for hovering and descent operations when dealing with a nearly empty first stage.
Take the Atlas V for example. A standard Atlas V 401 configuration has a GLOW of about 340 tonnes, but the empty first stage weighs only 22.5 tonnes, and is powered by only one RD-180 engine (which has two nozzles). The RD-180 generates 4 meganewtons of thrust, so when the first stage is nearly dry you’ll end up with almost 18 gees of acceleration. Imagine trying to do a precision landing with that sort of thrust in play, it’s just not practical. You need something closer to 1 gee, and there’s no way you can throttle an RD-180 down to 6% thrust. The Delta IV also only has one engine per first stage core, so it’s in the same boat.
MikeR, I think Robin, Edward and M Puckett explained it adequately?
Yes, thank you all – that was a big help!
Thanks.
It is always nice to get knowledgeable information that you cant glean from wikipedia or the popular press.
Probably worthwhile to bring up the recent Long March launch failure, taking with it a quarter billion dollar Brazillian satellite. Pretty bad timing for them to try to prevent so much of their commercial launch business fleeing to SpaceX.
Question: How much of the re-usability of a F9 is determined by the hardware and how much by the software?
From a hardware standpoint it may be harder for the dinosaurs to catch up but the barriers might be much lower on the software end.
When SpaceX did their first stage test a couple launches back, it was a big deal that they tried to restart the first stage and how this hadn’t been tried before by other companies. But my question is why? Is it because of the limitations of the hardware or is it because the software hadn’t progressed far enough to allow it?
I suspect a major reason why, is that they didn’t need to in order to fulfill contract requirements. The government was going to buy those launches and the companies would make a nice profit regardless so why go the extra steps for no additional compensation? That is the type of thinking that comes out of MBA finance classes as opposed to the more entrepreneurial approach of SpaceX.
I might be off base here someplace but with SpaceX demonstrating that it can be done, other companies might follow suit and find that it is easier to do today than it was twenty or thirty years ago, especially if there are not major limitations in adapting the hardware.
That is the type of thinking that comes out of MBA finance classes as opposed to the more entrepreneurial approach of SpaceX.
That’s not my experience of MBAs. Maintaining a competitive edge is a necessary fact of every business. The failure of some parts of the launch industry to innovate is a result of government interference, nothing more. In the areas of the industry where the government has traditionally had less interest, we’ve seen greater innovation, and in the areas of the industry where the government is the primary customer, we’ve had very little. This has nothing to do with “MBA finance classes”.
Many people inside ULA, for example, have proposed reusability experiments which would cost very little and introduce very little risk to operations. For example, I remember reading about some CORONA-style in-air recovery techniques being proposed for Delta core stage recovery. Why didn’t it happen? Simply: ULA isn’t free to spend their own money. This is a result of the consent agreement that was put in place to allow the merging of the two launch vehicle efforts of Boeing and Lockheed Martin. It was put in place to ensure dissimilar access to space for the US government. Innovation has been systematically stifled. No amount of entrepreneurial spirit at ULA can change this.
I’ve been pondering the possibility of filing a suit with the FTC to force a divestiture of ULA from Boeing/Lockmart.
Overdue. A recent example of the pressure from the parent companies on ULA was “Inspiration Mars” imploding with the “necessity” of using SLS. Though from my sources, Tito was dedicated to SLS, maybe he would have felt _he_ didn’t need it if ULA had been able to provide an alternative in spirited fashion.
Then there’s the despicable story from a year or two back concerning a certain former Congressperson…
Rand, Charles,
I don’t know how likely winning a suit like that would be, but it would be neat to see what ULA could do to compete if it were privately held by someone other than Boeing and LM. They’d still have to deal with a semi-neurotic main customer (NRO), and would need to have the funds to do innovative development. But they definitely have the creativity and talent to give SpaceX a run for its money if they weren’t chained down by two parent companies that often directly compete with them.
~Jon
“That’s not my experience of MBAs. Maintaining a competitive edge is a necessary fact of every business. ”
Not necessarily. Until recently ULA has had a virtual monopoly on government launches. This means there is no pressure or incentive to have a competitive edge. There was hardly even any competition that would determine what a competitive edge is. And because the government is their customer, they wont make changes to their launchers unless the government wants them to. So while that is in part due to the role of government there is also a role played by business philosophy in making these decisions.
Any new capital project would have to meet an IRR and then compete against all of the other projects that also beat the hurdle. A company with a government monopoly has a guaranteed profit and it is stable. It wouldn’t make fiscal sense to be developing re-usable technology because it represents a huge capital investment without any readily apparent returns or more importantly, returns greater than the alternatives.
“Many people inside ULA, for example, have proposed reusability experiments which would cost very little and introduce very little risk to operations.”
And those programs would have to go threw a process that weighs their ability to make profit over other proposed projects and existing projects. I am not familiar with the project you are talking about. It may very well have turned out to be inexpensive but the process to get to implementation sure wouldn’t have been cheap. And why even bother if the customer doesn’t want it or it isn’t part of the current deal?
I bet ULA would be more than happy to build re-usable rockets if the government paid them too.
“Why didn’t it happen? Simply: ULA isn’t free to spend their own money.”
I am also not familiar with the consent agreement that you mention but there is someone who does tell them how to spend their money. And in the case of restricting ULA, it is probably to protect the parent companies.
So, I agree that the government plays a large roll but I also think business does as well. Business people have to operate in the environment created for them, especially if your customer is the government. Sometimes that means maintaining the status quo as a means to preserve steady profits.
This is why I am hopeful for SpaceX. Maybe they can shake things up and get these older aerospace companies taking some risks and building things out of competition instead of merely fulfilling cost plus contract requirements.
I suspect that re-usability such that SpaceX is attempting is really something that needs to be designed in from the beginning. There must be 10,000 things that need to be accommodated in the booster design to even begin to attempt a soft landing — the ability to restart the engine just a small portion of those things.
The thing about the dinosaurs is they have deep pockets and the engineering talent to create competition for a reusable Falcon 9. What’s unknown is whether they have the will.
I liken this to Henry Ford using the assembly line to mass produce automobiles. How much of his contemporary competition simply pulled up stakes and went out of business, and how many learned from observing what Ford was doing and quickly game him competition.
If you plan to land on land, I suspect that’s true. But there were reasonably well-developed plans to try reusing the first stage of the Saturn V, which certainly wasn’t designed for it; the problem was that it made no financial sense unless it was going to make at least a few dozen flights.
Plus, realistically, given it was going to land in the sea, I suspect they’d need a bunch of flights just to figure out how to protect it from the sea water well enough to fly again.
However, I suspect most of us would like to see real competition for SpaceX. We just don’t think the existing launch companies have much chance of doing so.
In fact the last (essentially) fully reusable design in the (downward) evolution of Shuttle from optimal reusable systems use a fly-back ( yes, winged) version of the Saturn V first stage.
When that was rejected because of costs, they went to an external tank/strapon booster configuration, except the strapons were liquids and had clamshell doors designed to protect the motors from seawater. They were intended to be reused, but who knows how practical it would have been.
Finally, when _that_ design was rejected because the peak year development budget (in early ’70’s dollars) was above $1billion, they went to the solids, and convinced themselves that they could be cheaply reusable. And safe.
We all know how that turned out.
“I suspect that re-usability such that SpaceX is attempting is really something that needs to be designed in from the beginning. There must be 10,000 things that need to be accommodated in the booster design to even begin to attempt a soft landing — the ability to restart the engine just a small portion of those things.”
That would speak to how forward looking SpaceX is. Perhaps with SpaceX leading the way, other companies will find it isn’t as hard as they thought it was thirty years ago.
Much of it is dependent on hardware. You need to have an engine that you can restart, which is rarely the case for first stages, you need to have the right engine configuration to be able to provide thrust in the right range to facilitate hovering and also have enough control authority for targeted re-entry and landing, you need the stage to be strong enough to survive entry and descent.
That’s why the F9 v1.1 use TEA-TEB igniters, why it uses the “octo-web” engine configuration, why the central engine is gimballed, and why it has supports and mount points for landing legs. It is simply not the case, not by a long shot, that you could update the software of an Atlas V, Delta IV, Soyuz, or Ariane 5 and enable an RTLS flight profile for the first stage, even aside from the landing gear issues.
The Falcon 9 v1.1 / Falcon 9-R was engineered from the ground up for reusability, the fact that it makes a dandy expendable first stage as well is a double use benefit that proves the genius of SpaceX. They are going to be cranking out v1.1 stages from their factory, paid for by the customers of their expendable launch services. Once they’ve validated the reusable flight-profile of the v1.1 stage they’ll already have hardware on hand, they’ll already have some flights they can reuse the stage from, and they’ll already have their manufacturing fully geared up for production. It’s a shockingly smart business play, and it’s almost certain to disrupt the entire spaceflight industry almost overnight unless some unknown calamity happens to the company.
“That’s why the F9 v1.1 use TEA-TEB igniters, why it uses the “octo-web” engine configuration, why the central engine is gimballed, and why it has supports and mount points for landing legs. It is simply not the case, not by a long shot, that you could update the software of an Atlas V, Delta IV, Soyuz, or Ariane 5 and enable an RTLS flight profile for the first stage, even aside from the landing gear issues.”
Excellent, thanks!
“The Falcon 9 v1.1 / Falcon 9-R was engineered from the ground up for reusability, the fact that it makes a dandy expendable first stage as well is a double use benefit that proves the genius of SpaceX.”
Yes, very impressive and it should go a long way in removing any doubts that they are running the company as some hobby shop out of a dude’s basement.
” It’s a shockingly smart business play, and it’s almost certain to disrupt the entire spaceflight industry almost overnight unless some unknown calamity happens to the company.”
I would add the disruption comes not just from the re-usability but through their pricing strategy. It would have been very easy to undercut their competitors by $10m and rake in the contracts instead of offering drastically lower prices.
The great thing about SpaceX is that they aren’t just a company who builds rockets as an excuse to make money, they’re a company that makes money as an excuse to make rockets. That sort of passion has lots of follow through benefits that puts them well ahead of the game but also gives them market advantage (the way any creator with a legitimate passion for the field they’re in has).
If Musk merely wanted to make a profit of an average of, say, a billion a year over the next decade he could easily do that without reusability, without the MCT, without the manned Dragon. But it’s those things which will propel SpaceX even farther ahead of its competitors over the next several years, and in large part they are built out of a sense of vision and imperative.
For starters having 9 first stage engines means that the throttling requirements are much less. You really only need one throttleable engine. Try that with an Atlas 5.
I know there are other methods than a vertical powered landing.
I didn’t follow this. Could you explain it? – Thanks.
Not sure how many engines Delta has, but I believe the point is that if, say, you need 10% thrust to land, a stage with one engine would have to throttle it down to 10%, while a stage with nine engines would only have to throttle down to 90%.
When you have multiple engines, you can shut the unnecessary ones down.
Almost all of the first stage mass is fuel. Engines capable of >1g acceleration of a fully fueled first stage plus the upper stage(s) plus the payload will have far, far too much thrust for hovering and descent operations when dealing with a nearly empty first stage.
Take the Atlas V for example. A standard Atlas V 401 configuration has a GLOW of about 340 tonnes, but the empty first stage weighs only 22.5 tonnes, and is powered by only one RD-180 engine (which has two nozzles). The RD-180 generates 4 meganewtons of thrust, so when the first stage is nearly dry you’ll end up with almost 18 gees of acceleration. Imagine trying to do a precision landing with that sort of thrust in play, it’s just not practical. You need something closer to 1 gee, and there’s no way you can throttle an RD-180 down to 6% thrust. The Delta IV also only has one engine per first stage core, so it’s in the same boat.
MikeR, I think Robin, Edward and M Puckett explained it adequately?
Yes, thank you all – that was a big help!
Thanks.
It is always nice to get knowledgeable information that you cant glean from wikipedia or the popular press.
Probably worthwhile to bring up the recent Long March launch failure, taking with it a quarter billion dollar Brazillian satellite. Pretty bad timing for them to try to prevent so much of their commercial launch business fleeing to SpaceX.