-
What makes an Atlas V expensive?
by
LouScheffer
on 06 May, 2014 01:29
-
Where does the expense come from in the Atlas V? We know from recent disclosures that the RD-180 costs about $10M each (at least in the first order of engines). That can't be too different from 9 Merlin engines. The RL-10 *is* expensive, but even an expensive engine should be only $20M or so, not nearly enough to account for the difference in cost. Avionics can't be that big of a fraction, and it's hard to see how tanks can be hundreds of million more each. So where does the money go? Why can't you launch an Atlas V for $150M or so?
-
#1
by
cleonard
on 06 May, 2014 01:48
-
The bottom line is billed man hours. Even raw material prices have a lot of labor hours in them.
The RL10 is a classic example. It takes a lot of time to build.
-
#2
by
Avron
on 06 May, 2014 01:53
-
big pockets
-
#3
by
arachnitect
on 06 May, 2014 22:41
-
-Older workforce gets paid more, costs more to insure.
-More infrastructure: 2 launch pads, each with VIF. Liquid Hydrogen.
-Solids handling and markup
-support for multiple configurations
-Gov't requirements.
Keep in mind that really the only time we see the "cost" of an Atlas V it's the total value of a launch service contract to a gov't agency that includes a lot of services that aren't strictly the cost of the rocket. NASA or DOD almost always order "all the fixings."
The total cost value for the MAVEN launch service is approximately $187 million. This estimated cost includes the task ordered launch service for the Atlas plus additional services under other contracts for payload processing; launch vehicle integration; mission unique launch site ground support; and tracking, data and telemetry services.
So building the rocket, putting it on the pad, and pressing "GO" probably does cost less than $150M, but all the other stuff adds up.
Avionics can't be that big of a fraction,
How much did it cost to write the software that runs on those avionics?
-
#4
by
edkyle99
on 07 May, 2014 19:13
-
Here's what I've long believed about EELV costs in general.
1. The program was designed from the outset to cover a huge payload range (from Delta II to Titan IV Centaur) using one vehicle. That's a noble goal, but it means that a launch site that's handling a 1.2 tonne DMSP satellite is also designed to host massive Heavy payloads weighing 17 tonnes or more, and the launch vehicles for both. The same is true of the factory and the processing facilities and the personnel needs and the rockets themselves. The result is a compromise program that may cut launch costs for a few Heavy payloads, but that also raises costs for small payloads. I'm sure someone in the Pentagon has a spreadsheet that proves this approach lowers overall costs. ...
2. The original program called for ONE winning vehicle. The Pentagon decided to keep TWO vehicles. That decision alone DOUBLED costs. One reason that Atlas costs so much, then, is that those dollars are also helping to pay for Delta IV Heavy infrastructure, for two (4 and 5 meter) Delta IV upper stage production lines, Delta IV launch pads, Delta IV strap on booster contracts, Delta IV payload fairings, Delta IV avionics, Delta IV RL10 production, RS-68, etc. and etc..
- Ed Kyle
-
#5
by
BrightLight
on 07 May, 2014 19:27
-
One of the issues with the production of anything is the number of people required to make the widget - that being said, how;
many people are required to keep the Atlas V system in production?
Are the same people used to keep the Delta V going? and
Does the Atlas V cost have to include the costs for the Delta IV?
-
#6
by
JBF
on 07 May, 2014 19:46
-
Without an impetus to streamline labor, management, manufacturing, and integration - virtually every company will mismanage costs since there's no penalty for doing so. It's more human nature than "oh! they should have subcontracted that component manufacturer instead."
3 fuel types? If someone was designing a rocket on paper from scratch, I tend to think they wouldn't design each stage with a different propellant. Those sort of decisions add up over the whole project and lead to ballooning costs. This is largely due to the making decision based on political will, rather than engineering savvy. If you want congressional support for some big DoD project, they serve up subprojects as political sweeteners for congressional support in different states.
It's not one simple reason, but a general culture that doesn't value streamlined and efficient engineering as much as garnering political capital.
Some people are going to flip out at me for the real politik, but who can deny that's how it works? They're still good engineers, just not as good as they could be when they take orders from politicians.
The three different engines comes from optimizing for performance not cost. Solids are really great for the initial push off the pad. RP-1 is great for lower atmosphere and LH2 is great for upper and space.
-
#7
by
JBF
on 07 May, 2014 19:51
-
The three different engines comes from optimizing for performance not cost. Solids are really great for the initial push off the pad. RP-1 is great for lower atmosphere and LH2 is great for upper and space.
That may be true, but wouldn't it sometimes better to sacrifice some of the performance for cheaper and simpler manufacturing and integration? Who cares if you've got the most powerful and high performance engines/rocket in the world if no one can afford it.
No argument from me, I approve of what SpaceX is doing. However that is not what has occurred historically. For a long time we have been optimizing for performance and reliability above all else.
-
#8
by
zt
on 07 May, 2014 19:56
-
Does anyone in Congress, DoD, USAF, NRO, etc. have a breakdown of price for each of those line items?
| Line Item | Delta IV | Atlas V |
| yearly first stage contract/maintaining production line | $?M | $?M |
| yearly first stage engine contract/maintaining production line | $?M | $?M |
| yearly second stage contract/maintaining production line | $?M | $?M |
| yearly second stage engine contract/maintaining production line | $?M | $?M |
| yearly strap on booster contract/maintaining production line | $?M | $?M |
| yearly payload fairings contract/maintaining production line | $?M | $?M |
| yearly avionics contract/maintaining production line | $?M | $?M |
| yearly west coast pad upkeep | $?M | $?M |
| yearly east coast pad upkeep | $?M | $?M |
| yearly infrastructure contract | $?M | $?M |
| etc. | $?M | $?M |
Then you can say that the price of maintaining the infrastructure buys you one complete Delta IV and one complete Atlas V core (first and second stages and fairing) per year, which they would have to build in order to not forget how and demonstrate that they indeed did not forget how. Actually launching this "free" core costs extra. Building more than one per year costs extra.
And then the price of a launch should only include the things it actually uses just for that launch. Needs a core because it's not the first one in a year? Extra. Needs the fancier upper stage? Extra. Needs strap on boosters? Extra. Needs integration with new payload type? Extra. Needs payload processing? Extra. Special trajectory/handling/secrecy/tracking? Extra.
So the price of a launch is separate from the yearly "maintaining the infrastructure" price, and the infrastructure upkeep is per type of rocket.
Even if each line item is cost-plus, at least the buyer knows what they're paying for.
-
#9
by
Jim
on 07 May, 2014 20:29
-
One of the issues with the production of anything is the number of people required to make the widget - that being said, how;
many people are required to keep the Atlas V system in production?
Are the same people used to keep the Delta V going? and
Does the Atlas V cost have to include the costs for the Delta IV?
The production, engineering and operations teams cross product lines.
-
#10
by
Jim
on 07 May, 2014 20:35
-
The three different engines comes from optimizing for performance not cost. Solids are really great for the initial push off the pad. RP-1 is great for lower atmosphere and LH2 is great for upper and space.
The original EELV concepts did not call for solids. It was the commercial market and growth of comsats that drove the use of solids. The CCB and VIF were basically designed before the changes that is why the asymmetric SRB attachments and why the VIF has level numbers X.X in between whole numbers. Otherwise, the Atlas V was an Evolved ELV, it took elements of Atlas and Titan.
The small EELV was for GPS, DMSP (Delta II, Titan II)
the medium for DSCS, UFO, etc ( Atlas II)
The heavy was for DSP, Milstar and NRO sats (Titan IV)
The comsat market changed, the use of commercial comsat buses and termination of the small EELV changed the landscape. The addition of solids became the intermediate class and the bulk of the DOD spacecraft are now in that range.
-
#11
by
Jim
on 07 May, 2014 20:45
-
Does anyone in Congress, DoD, USAF, NRO, etc. have a breakdown of price for each of those line items?
| Line Item | Delta IV | Atlas V |
| yearly first stage contract/maintaining production line | $?M | $?M |
| yearly first stage engine contract/maintaining production line | $?M | $?M |
| yearly second stage contract/maintaining production line | $?M | $?M |
| yearly second stage engine contract/maintaining production line | $?M | $?M |
| yearly strap on booster contract/maintaining production line | $?M | $?M |
| yearly payload fairings contract/maintaining production line | $?M | $?M |
| yearly avionics contract/maintaining production line | $?M | $?M |
| yearly west coast pad upkeep | $?M | $?M |
| yearly east coast pad upkeep | $?M | $?M |
| yearly infrastructure contract | $?M | $?M |
| etc. | $?M | $?M |
Then you can say that the price of maintaining the infrastructure buys you one complete Delta IV and one complete Atlas V core (first and second stages and fairing) per year, which they would have to build in order to not forget how and demonstrate that they indeed did not forget how. Actually launching this "free" core costs extra. Building more than one per year costs extra.
And then the price of a launch should only include the things it actually uses just for that launch. Needs a core because it's not the first one in a year? Extra. Needs the fancier upper stage? Extra. Needs strap on boosters? Extra. Needs integration with new payload type? Extra. Needs payload processing? Extra. Special trajectory/handling/secrecy/tracking? Extra.
So the price of a launch is separate from the yearly "maintaining the infrastructure" price, and the infrastructure upkeep is per type of rocket.
Even if each line item is cost-plus, at least the buyer knows what they're paying for.
1. Those numbers are proprietary
2. DOD doesn't buy rockets, it buys fix price launch services. So the DOD does not deal with component costs.
-
#12
by
veblen
on 07 May, 2014 20:59
-
Without an impetus to streamline labor, management, manufacturing, and integration - virtually every company will mismanage costs since there's no penalty for doing so. It's more human nature than "oh! they should have subcontracted that component manufacturer instead."
3 fuel types? If someone was designing a rocket on paper from scratch, I tend to think they wouldn't design each stage with a different propellant. Those sort of decisions add up over the whole project and lead to ballooning costs. This is largely due to the making decision based on political will, rather than engineering savvy. If you want congressional support for some big DoD project, they serve up subprojects as political sweeteners for congressional support in different states.
It's not one simple reason, but a general culture that doesn't value streamlined and efficient engineering as much as garnering political capital.
Some people are going to flip out at me for the real politik, but who can deny that's how it works? They're still good engineers, just not as good as they could be when they take orders from politicians.
The three different engines comes from optimizing for performance not cost. Solids are really great for the initial push off the pad. RP-1 is great for lower atmosphere and LH2 is great for upper and space.
That may be true, but wouldn't it sometimes be better to sacrifice some of the performance for cheaper and simpler manufacturing and integration? Who cares if you've got the most powerful and high performance engines/rocket in the world if no one can afford it.
By the same token if a cheap rocket cannot do the mission the price is irrelevant.
-
#13
by
zt
on 07 May, 2014 21:35
-
Does anyone in Congress, DoD, USAF, NRO, etc. have a breakdown of price for each of those line items?
...
1. Those numbers are proprietary
2. DOD doesn't buy rockets, it buys fix price launch services. So the DOD does not deal with component costs.
1. It's ok that
I don't get these numbers, as long as the buyer has them. But the only way to know why the price is high is to find out the real component costs.
2. ULA mentioned the government not having insight into SpaceX costs. That meant to me that the government
does have insight into ULA's costs. What's the difference between the insight the government has into ULA vs. SpaceX?
3. For a sole-source contract for a service critical national security that costs billions, the government should demand total transparency. You can use the free market to control efficiency if you have a market, which implies multiple providers not colluding. When only a single entity is capable of building such rockets (before SpaceX is certified), the only way to control efficiency is to have total transparency. Does buying navy ships work the same way? Navy knows the total price, not the component cost? If so, that's nuts.
-
#14
by
baldusi
on 07 May, 2014 22:53
-
One thing to consider is the fact that DoD's requirements made optimizations that increased cost just enough to price them out of the market. The recent commercial wins for Atlas V and NASA's performance shows that they are a bit higher, not ridiculously higher. If they had down selected to one, the economies of scale alone might have allowed costs to be competitive. Of course that with down select to one there would be very few slots for commercial launches.
-
#15
by
Jim
on 07 May, 2014 23:00
-
1. It's ok that I don't get these numbers, as long as the buyer has them. But the only way to know why the price is high is to find out the real component costs.
2. ULA mentioned the government not having insight into SpaceX costs. That meant to me that the government does have insight into ULA's costs. What's the difference between the insight the government has into ULA vs. SpaceX?
3. For a sole-source contract for a service critical national security that costs billions, the government should demand total transparency. You can use the free market to control efficiency if you have a market, which implies multiple providers not colluding. When only a single entity is capable of building such rockets (before SpaceX is certified), the only way to control efficiency is to have total transparency. Does buying navy ships work the same way? Navy knows the total price, not the component cost? If so, that's nuts.
1. Neither does the buyer because it is buying launch services.
2. The govt has ULA on contract, it doesn't have Spacex on contract.
3. Apples and oranges. Again, gov't is buying launch services and not a rocket. The Navy doesn't buy carrier services but a carrier ship.
-
#16
by
TomH
on 07 May, 2014 23:21
-
Liquid Hydrogen.
Atlas V uses RP-1, not LH
2
-
#17
by
baldusi
on 07 May, 2014 23:23
-
Liquid Hydrogen.
Atlas V uses RP-1, not LH2
Centaur does.
-
#18
by
edkyle99
on 08 May, 2014 01:59
-
Liquid Hydrogen.
Atlas V uses RP-1, not LH2
Centaur does.
Yes, but in pretty much the right spot to provide the most performance for the dollar spent. An RP/LOX "Centaur" would boost less than half as much to GTO on an Atlas V.
- Ed Kyle
-
#19
by
baldusi
on 08 May, 2014 14:23
-
Liquid Hydrogen.
Atlas V uses RP-1, not LH2
Centaur does.
Yes, but in pretty much the right spot to provide the most performance for the dollar spent. An RP/LOX "Centaur" would boost less than half as much to GTO on an Atlas V.
- Ed Kyle
The original question was on the use of multiple propellents, and arachnitect stated that Atlas V used RP-1, hence my answer.
Regarding the economic optimality, that's a more difficult question. Is Atlas V cheaper than an USA made Zenit? I don't really know.
-
#20
by
edkyle99
on 08 May, 2014 15:49
-
Regarding the economic optimality, that's a more difficult question. Is Atlas V cheaper than an USA made Zenit? I don't really know.
I guess the answer might be addressed hypothetically by asking another question. Would it cost more to double the number of RD-180s and to have an extra RP upper stage or two, or to maintain the existing RL10 setup?
The nearest comparison is Falcon 9, but v1.1 doesn't quite lift as much as even a 401 to GTO, and I doubt that we're seeing the long-term price point for SpaceX launch services.
- Ed Kyle
-
#21
by
baldusi
on 08 May, 2014 16:16
-
Regarding the economic optimality, that's a more difficult question. Is Atlas V cheaper than an USA made Zenit? I don't really know.
I guess the answer might be addressed hypothetically by asking another question. Would it cost more to double the number of RD-180s and to have an extra RP upper stage or two, or to maintain the existing RL10 setup?
The nearest comparison is Falcon 9, but v1.1 doesn't quite lift as much as even a 401 to GTO, and I doubt that we're seeing the long-term price point for SpaceX launch services.
- Ed Kyle
F9 v1.1 does 3650kg to 1,500m/s GTO, 401 does 3,460kg. But a Zenit would have a single RD-171M always. No solids, because the core already has a T/W of 2 (core alone). And a high T/W SC RP-1/LOX second stage is used for LEO and smal GTO, you can add a third stage for higher performance GTO or a heavy.
The NWO was quoted 10M per solid on each mission. I don't believe that RD-171M is more than 10M more expensive than an RD-180. And you get the equivalent to 431 or even 541 performance to US staging point (Centaur is more efficient, yes). But then you have the heavy.
So you'd still have three stages (core, second and third), and less performance steps. But also you'd have a single propellant, bigger economies of scale with engines and cores, and you could have done something like the Falcon 9 flow with erector and an optional MST for government payloads. It's not cheaper if you think only of DoD specs. But you can lower the costs for commercial and NASA.
-
#22
by
BrightLight
on 08 May, 2014 16:26
-
One of the issues with the production of anything is the number of people required to make the widget - that being said, how;
many people are required to keep the Atlas V system in production?
Are the same people used to keep the Delta V going? and
Does the Atlas V cost have to include the costs for the Delta IV?
The production, engineering and operations teams cross product lines.
If the Delta IV were removed, then some of these people can be removed from the system, reducing overhead cost?
-
#23
by
Jim
on 08 May, 2014 16:35
-
yes, and that would also include shutting down two launch pads, a HIF, a control center and offices, possibly a ship, etc
-
#24
by
edkyle99
on 08 May, 2014 16:41
-
F9 v1.1 does 3650kg to 1,500m/s GTO, 401 does 3,460kg.
Can you show me where that Falcon 9 number came from? I suspect it is an old number. SpaceX hasn't release details like that about v1.1 to my knowledge.
- Ed Kyle
-
#25
by
baldusi
on 08 May, 2014 20:14
-
F9 v1.1 does 3650kg to 1,500m/s GTO, 401 does 3,460kg.
Can you show me where that Falcon 9 number came from? I suspect it is an old number. SpaceX hasn't release details like that about v1.1 to my knowledge.
- Ed Kyle
I took the number from the NLS II site (185km x 35768km x 15deg). But it's also true that after the Thaicom mission SpaceX disclosed that they had reserved 300kg of GTO performance. Thus, to a 1,800m/s GTO, it can actually launch 5,150kg. Which is what SES contracted for, BTW. And this would agree with 3,650kg to a 1,500m/s GTO.
-
#26
by
BrightLight
on 08 May, 2014 20:17
-
yes, and that would also include shutting down two launch pads, a HIF, a control center and offices, possibly a ship, etc
In general terms - is this a whole lot of money or trivial?
roughly how many people are employed by ULA?
-
#27
by
baldusi
on 08 May, 2014 21:28
-
yes, and that would also include shutting down two launch pads, a HIF, a control center and offices, possibly a ship, etc
Control center and offices aren't going to be closed anyways because of Common Avionics?
-
#28
by
Jim
on 08 May, 2014 22:47
-
yes, and that would also include shutting down two launch pads, a HIF, a control center and offices, possibly a ship, etc
In general terms - is this a whole lot of money or trivial?
roughly how many people are employed by ULA?
3700
-
#29
by
Jim
on 08 May, 2014 22:48
-
Control center and offices aren't going to be closed anyways because of Common Avionics?
Delta and Atlas have separate LCC's.
-
#30
by
BrightLight
on 08 May, 2014 22:52
-
yes, and that would also include shutting down two launch pads, a HIF, a control center and offices, possibly a ship, etc
In general terms - is this a whole lot of money or trivial?
roughly how many people are employed by ULA?
3700
wow - my arm waving here - if ULA made a 25% reduction in force to only fly the Atlas and a average full time equivalent cost $250,000 per year, that's over $230 million - that can be a substantial portion of the expense in the Atlas.
-
#31
by
clongton
on 08 May, 2014 23:55
-
yes, and that would also include shutting down two launch pads, a HIF, a control center and offices, possibly a ship, etc
In general terms - is this a whole lot of money or trivial?
roughly how many people are employed by ULA?
3700
wow - my arm waving here - if ULA made a 25% reduction in force to only fly the Atlas and a average full time equivalent cost $250,000 per year, that's over $230 million - that can be a substantial portion of the expense in the Atlas.
You can't just cut percentages. There are skill sets that MUST be kept intact regardless of the workforce size.
-
#32
by
gospacex
on 09 May, 2014 00:21
-
Where does the expense come from in the Atlas V?
From ULA.
Gotta go....
-
#33
by
kevin-rf
on 09 May, 2014 11:53
-
I'll make the argument all launchers have subsidies, it is just ULA's are more visible.
Do SpaceX's prices include the full accounting costs it has taken to develop the Falcon 9? At $50 something million a flight, how long will it take to pay off the development costs?
Ariane had many subsidies in the past allowing lower current costs. Notice without these subsidies SpaceX is giving them a run for the money.
Proton was paid for and written off long before it became available on the commercial market. Does Proton still use conscripts as part of the launch operations?
I think a handful of things keep the perceived costs so high.
1. Fully audit-able cost accounting.
2. A large well paid workforce. Putting all the employees on conscript gruel and wages would not fly.
3. Double checking/documenting everything (That is not free). An alligator waddles into the VIF, someone gets paid to write a report.
4. Being Horizontally integrated adds cost at every step of the process. Instead of paying the overhead of a single company, you are paying the overhead of all the suppliers plus ULA. SpaceX by being vertically integrates avoids having to pay this additional overhead.
-
#34
by
WHAP
on 09 May, 2014 13:24
-
I'll make the argument all launchers have subsidies, it is just ULA's are more visible.
Do SpaceX's prices include the full accounting costs it has taken to develop the Falcon 9? At $50 something million a flight, how long will it take to pay off the development costs?
Ariane had many subsidies in the past allowing lower current costs. Notice without these subsidies SpaceX is giving them a run for the money.
Proton was paid for and written off long before it became available on the commercial market. Does Proton still use conscripts as part of the launch operations?
I think a handful of things keep the perceived costs so high.
1. Fully audit-able cost accounting.
2. A large well paid workforce. Putting all the employees on conscript gruel and wages would not fly.
3. Double checking/documenting everything (That is not free). An alligator waddles into the VIF, someone gets paid to write a report.
4. Being Horizontally integrated adds cost at every step of the process. Instead of paying the overhead of a single company, you are paying the overhead of all the suppliers plus ULA. SpaceX by being vertically integrates avoids having to pay this additional overhead.
5. Profits. Having two masters that want profits (documented in the LMT and BA annual report), vs. one who wants none. And horizontal integration also increases cost beyond overhead, as each supplier is going to make money. These aren't profits of 20% at each stage, but they still add up.
-
#35
by
edkyle99
on 09 May, 2014 13:39
-
F9 v1.1 does 3650kg to 1,500m/s GTO, 401 does 3,460kg.
Can you show me where that Falcon 9 number came from? I suspect it is an old number. SpaceX hasn't release details like that about v1.1 to my knowledge.
- Ed Kyle
I took the number from the NLS II site (185km x 35768km x 15deg). But it's also true that after the Thaicom mission SpaceX disclosed that they had reserved 300kg of GTO performance. Thus, to a 1,800m/s GTO, it can actually launch 5,150kg. Which is what SES contracted for, BTW. And this would agree with 3,650kg to a 1,500m/s GTO.
We'll have to see about those NLS II numbers. They seem overoptimistic, or simply outdated, to me.
- Ed Kyle
-
#36
by
Roy_H
on 16 May, 2014 17:00
-
I'll make the argument all launchers have subsidies, it is just ULA's are more visible.
Do SpaceX's prices include the full accounting costs it has taken to develop the Falcon 9? At $50 something million a flight, how long will it take to pay off the development costs?
Ariane had many subsidies in the past allowing lower current costs. Notice without these subsidies SpaceX is giving them a run for the money.
Proton was paid for and written off long before it became available on the commercial market. Does Proton still use conscripts as part of the launch operations?
I think a handful of things keep the perceived costs so high.
1. Fully audit-able cost accounting.
2. A large well paid workforce. Putting all the employees on conscript gruel and wages would not fly.
3. Double checking/documenting everything (That is not free). An alligator waddles into the VIF, someone gets paid to write a report.
4. Being Horizontally integrated adds cost at every step of the process. Instead of paying the overhead of a single company, you are paying the overhead of all the suppliers plus ULA. SpaceX by being vertically integrates avoids having to pay this additional overhead.
5. Profits. Having two masters that want profits (documented in the LMT and BA annual report), vs. one who wants none. And horizontal integration also increases cost beyond overhead, as each supplier is going to make money. These aren't profits of 20% at each stage, but they still add up.
Elaborating on point 4. Someone said that ULA reported an average profit of $43M per launch. Elon stated that ULA has in extreme cases 5 sub-contractor deep structure. If each of these takes a similar percentage markup, the costs can easily double. This is not all ULA's fault, the politics of the business is that to get Congressional support, everybody wants a piece of the pie, and wants some parts sub-contracted to their district. Congressmen want to keep these jobs in their district, and will fight SpaceX at every turn.
ULA will never be able to compete with SpaceX without a more vertical integration, but they won't have to if Congress has any say.
-
#37
by
Go4TLI
on 16 May, 2014 17:25
-
5. Profits. Having two masters that want profits (documented in the LMT and BA annual report), vs. one who wants none. And horizontal integration also increases cost beyond overhead, as each supplier is going to make money. These aren't profits of 20% at each stage, but they still add up.
If this is serious it shows the ridiculousness of some of the SpaceX crowd.
Yes, now Musk and SpaceX want no profits whatsoever. They are going to run a company that forever nets to zero dollars or loses money. They will do all these grand things they have spoken about from the money that grows on the tree in the backyard. In addition, all SpaceX employees will forever forgo any pay increases regardless if they have earned or the cost of living changes. Such wonderful people....
And "horizontal integration" gets a unfair pounding as if it is "bad" in the same way people rail away at certain types of contracts. It depends on the situation and it doesn't always make sense to establish a capability, gain the expertise, etc when a supplier of those products and that expertise is available.
Take Boeing for example. It would be very difficult as a company from a management, cost and competitive viewpoint to have every capability under the Boeing roof to produce everything ever needed for every single type of aircraft (commercial and military), sats of all kinds, spacecraft of all kinds and every other program they are involved in.
-
#38
by
Antares
on 16 May, 2014 18:14
-
How do you equate profits and pay increases? If SpaceX wants pay increases, it will raise prices. There's no reason to assume prices are static.
Do you consider revenue that goes back into R&D of new things to be profit? There's a reason why SpaceX doesn't IPO: it's the loss of control (and the openness that regulations require of a publicly held company).
Everyone said the same thing about vertical integration, until SpaceX did it. Seems to work. Assumptions have to be turned on their heads every once in a while.
-
#39
by
pathfinder_01
on 16 May, 2014 22:40
-
And "horizontal integration" gets a unfair pounding as if it is "bad" in the same way people rail away at certain types of contracts. It depends on the situation and it doesn't always make sense to establish a capability, gain the expertise, etc when a supplier of those products and that expertise is available.
Take Boeing for example. It would be very difficult as a company from a management, cost and competitive viewpoint to have every capability under the Boeing roof to produce everything ever needed for every single type of aircraft (commercial and military), sats of all kinds, spacecraft of all kinds and every other program they are involved in.
I think that in the rocket industry horizontal integration often does not add value. One of the reasons why horizontal integration works, is if the item or part has a larger market than just your own product. The cost per unit could be spread over far more units than you could produce yourself.
Jet engines for instance, aircraft often get their engines replaced. One model of jet engine can be used in more than a single model of airplane both civilian and military and thus having aircraft manufactures separate from engine manufactures works. Boeing makes the air frame and say GE makes the engine.
Another more likely example might be tires and automobiles. Both new and used autos need tires and thus there is a huge market for tires and tire companies like Firestone could offer cheaper tires to auto companies like Ford because they were selling tires to both new and used cars made by Ford as well as others.
Solar panels and satellites might be an third. There are more than one satellite manufacture(Boeing, Orbital, Lockheed Martin, and others each build satellites) and thus there is an market for ATK’s solar panels.
With rocket parts and rocket engines there really is not a huge market to spread costs over many units. ULA was until recently an monopoly. The RD-180 is only cheaper because it is not manufactured in the US. Which for a rocket meant to launch national security payloads saved some money, but just added risk. About the only engines that might be used across more than a single type of rocket are upper stage ones like the RL-10 or some of the solid fueled ones.
I think there may be many inefficiencies built into the standard model of rocket manufacture. The one that ULA uses that don’t make sense and simply add cost or risk. There are not enough rocket manufactures as well as rocket models to spread the cost over multiple units over and instead of save money it just adds costs with more overhead and supplier risk. ATK for instance is not the only solar panel manufacturer and it's panels could be replaced but replacing the RD-180 on Atlas is an major problem.
-
#40
by
baldusi
on 17 May, 2014 01:07
-
Errr, RS-68 shared a lot of infrastructure with RS-25. That went away. Then the contract negotiated right after the dissolution of the URSS needed to be extended and Energomash was in a much better position to negotiate. That explains most of first stage propulsion.
In SpaceX case, what makes sense for building in house is that they have core competencies in doing space related manufacturing. They have the experience, expertise, software and process control for the whole life cycle of space qualified products. In many cases, implementing that system is more expensive than the design and tooling of the part. Thus,while there isn't a big market for space qualified products, it will make sense for SpaceX to do a lot in house.
-
#41
by
Antares
on 17 May, 2014 15:04
-
SpaceX also has the benefit of designing both a vehicle and a plant from a clean sheet. The rocket and its parts could be designed for modern manufacturing since there were no designs already existing. Existing designs can complete development faster (and therefore at lower development cost) but possibly at the expense of recurring cost. Having no build tools means SOMETHING has to be bought, so the company can optimize for tools that can fabricate a lot of different things. Then again, ULA Decatur wasn't there 18 years ago.
-
#42
by
watermod
on 17 May, 2014 17:25
-
You are forgetting SpaceX "sister" companies.
Tesla - Electric vehicle, electro-motive, battery and very high-tech mass production skills it can share with SpaceX
SolarCity - experience in solar power generation, storage and distribution, sourcing and experimenting with cells from a wide variety of commercial sources. Lots of this applies to rockets, sats and bases.
Finance - Elon is not a "banker" anymore but he definitely knows the movers and shakers and ins and outs after creating and owning
Pay Pal. - a good knowledge of multiple sources and sinks of capital never hurts
Zip2 - This product let Elon explore using computers to map complex systems in unique ways (in Zip2's case cities)
Then there are his Hi-Tech investors and what they may or may not be bringing to the table. (Not just VCs)
I would consider it more appropriate to consider a profile of Musk's firms to be better matched to a Japanese Keiretsu than typical Western firms. The nearest US match would have been the Motorola combine under Galvin Sr in the 80s and 90s.
http://en.wikipedia.org/wiki/Keiretsu