6 ms·

NASA's problems with launch tower for SLS are getting worse

45 points by NavinF 2y ago

NavinF 2y ago
>$2.7 billion. Such a cost is nearly twice the funding it took to build the largest structure in the world, the Burj Khalifa, which is seven times taller. > NASA officials informed us they do not intend to request a fixed-price proposal from Bechtel
- gexla 2y ago
  It seems a significant addition to the cost overruns is labor. Maybe this is a good example of the cost reality of bringing manufacturing back to the US. Given a difficult timeline, the US workers would need to be putting in a lot of overtime. Many of the laborers who took part in building the Burj Khalifa were being paid less than what the US workers would spend on a meal. https://en.wikipedia.org/wiki/Burj_Khalifa https://en.wikipedia.org/wiki/Burj_Khalifa Edit: Here is the NASA report. https://oig.nasa.gov/office-of-inspector-general-oig/audit-reports/nasas-management-of-the-mobile-launcher-2-project/ https://oig.nasa.gov/office-of-inspector-general-oig/audit-r...
  - jiggawatts 2y ago
    > It seems a significant addition to the cost overruns is labor Only because they're billing for a ridiculous amount of time. Estimated completion time of 2027!? Wat? SpaceX has already built three and is planning more. By the time Bechtel completes their single tower, SpaceX will have half a dozen and a Moon colony. The contract should be torn up and the people that signed it barred from holding executive positions for life.
  - omoikane 2y ago
    The article seems to be using the cost of Burj Khalifa without accounting for inflation. $1.5 billion in 2009 would be about $2.2 billion in 2024. NASA's project is still expensive, but not quite twice as expensive as Burj Khalifa (unless the budget overruns again).
    - disillusioned 2y ago
      The OIG report is projecting it could be as much as $2.7 Billion all said and done.
thot_experiment 2y ago
Someone who follows this stuff, how does this compare to the starship catching towers? Both in terms of complexity and cost. My gut says the starship towers are about as complicated (in an abstract "does a bunch of tricky things wrt spaceship support" even if they're different things way) and I'm betting it's 100x? 1000x cheaper? I don't know anything though so if someone is more knowledgeable I would love to learn more.
- BHSPitMonkey 2y ago
  SpaceX doesn't tend to share that kind of information, so we may never know exactly. Plus, it's R&D they expect will amortize over the operational life of Starship rather than infrastructure being built for a narrow set of missions that I can't see resulting in much future reuse.
- glzone1 2y ago
  SpaceX built (and rebuilt) a tower practically overnight. They are building two more. They are building huge construction bays overnight. Their contractors are literally working nights. There is probably no comparison on cost or speed. ML-1 and ML-2 have been percolating along for over a decade. ML-1 is going to see just a few launches (3-4?). ML-2 is just 7 feet taller. Total between the two maybe $3.5 billion range? This is just for the SLS tower. Actual SLS / Orion / sustainment costs at the NASA centers are absolutely mind boggling. Then you compare that to what NASA gets with their contracts with SpaceX and the mind boggles.
- JumpCrisscross 2y ago
  > how does this compare to the starship catching towers? These are mobile while Starship’s towers are fixed. That alone massively increases the MLs’ complexity. Why are they mobile? I genuinely can’t tell beyond that’s how Apollo did it. A fixed tower with a moveable (or even temporary) structure built around it would have been simpler.
JCharante 2y ago
Why do governments allow cost follow pricing? Companies should bid a final price and pay for insurance that covers unexpected costs/delays.
- vimax 2y ago
  How do you expect insurance companies to create actuarial tables for one-off government R&D projects?
  - tedunangst 2y ago
    Insurance companies manage to find ways to price insurance for Hollywood actors storming off set.
  - NavinF 2y ago
    Just estimate the probability of payout. Yes you can do this for non-repeating, hard-to-model events: https://www.astralcodexten.com/p/in-continued-defense-of-non-frequentist https://www.astralcodexten.com/p/in-continued-defense-of-non... I do it all the time without making actuarial tables, as do insurance companies.
  - JCharante 2y ago
    Same way insurance companies insure satellite launches. Lots of guesses and getting another insurance company to underwrite their insurance deals.
ungreased0675 2y ago
NASA seems to be setting money on fire lately. It’s probably time to shutter the place and start over with the National Space and Air Administration.
- readthenotes1 2y ago
  About a decade ago and NASA employee worried to me that it was just becoming a white collar jobs program. Mission accomplished.
namlem 2y ago
SpaceX already built their launch/catch tower and didn't even bill the taxpayers for it.
glzone1 2y ago
Wild. But it's a pattern with these contracts. As a reminder, for SLS they decided to modify an exiting launch tower (ML-1). The original estimate to modify the tower was $54 million. Final cost will likely be $1 billion. And delayed of course. It's only going to get a few launches. You can see where this is going. Then they decided to do a new launch tower, ML-2. This tower is only 7 feet taller. They planned to incorporate lessons learned from the mess of ML-1 (haha). Started at $400M. Now going to be $2.7 billion (and delayed of course badly). These cost+ deals especially with multiple contractors are like rotting fruit that attracts maggots that glom on and provide negative value. Something that would take a day to a week elsewhere could be literally months. Every incentive is to move slowly. The contractor layers and paperwork needs on even basic changes are totally massive. The other thing that really jumps out is usually speed - these things will take FOREVER. I once even saw something about 8 track data recording in a govt contract, It's wild what is stacked into these things. The overhead to do anything can be wild. I once worked a govt job, and we needed an ipad for whatever reason (think something like foreflight on an ipad - most normal people would just buy an ipad to run foreflight). For all sorts of dumb and dumber reason, an ipad could not be purchased. I'd estimate $25K - $50K maybe went into trying to buy this damn ipad and coming up with workarounds. Can only go through central IT which wasn't even selected for price performance, has no customer service, and doesn't have apple SKU's. That's just the beginning. By the time you've wasted attorney time, contracting teams time on this ipad purchase and your own time, your managers time, all the downline folks time... the mind boggles. I think the final solution was to do a contract with a third party for a service, who could then buy the ipad and install the app on it as long as that part of the contract was less than some capital item threshold, and then that third party could enter into another agreement to make it available for their use. But the time to do this, then the time to come up with the agreement to use the ipad owned by this third party... total joke! I understood why teachers buy their own pencils. I'm sure a school can provide them, but a teacher probably doesn't want to deal with the headaches!
- disillusioned 2y ago
  Fire everyone at NASA involved in "project management" that allowed this to happen: >For the ML-2 project’s first 5 years, NASA lacked a reliable cost and schedule estimate, making it difficult for the Agency to accurately identify ML-2 budget requirements, be accountable to Congress and other stakeholders, and accurately measure project and contractor performance. Absolute dereliction of duty and complete incompetence. Bechtel has literally no _true_ incentive to deliver on time: the "awards" are folly and they've been awarded some of them even despite this abysmal performance. Instead, you can inflate the "cost" part of "cost plus" and just continue to suck on the teat of the taxpayer, while blaming supply chain and other challenges, all to not build what amounts to a several hundred foot tall elevator. (Obviously a gross simplification; I understand there are interconnected data, electrical, hydraulic, fluid, and other systems, along with being immediately adjacent an extremely high-thrust launch, but _still_.)
devoutsalsa 2y ago
If SLS is only getting a few launches, and we’re planning to establish a permanent presence on the moon, what will replace SLS? Starship and New Glenn?
- imtringued 2y ago
  Why are you so oddly fixated on SLS? SLS is a launch vehicle for Orion. What is needed is a replacement for Orion so NASA can ditch SLS. The only alternative to Orion is to use a capsule to LEO, transfer crew to a second fully fueled Starship, dock with HLS in NRHO, do the landing and come back, dock with the crew Starship and then do a deceleration burn, dock in LEO with the capsule and then return to earth. Why? Because the lunar Starship doesn't have enough delta V to return LEO and landing crew on earth via aerobraking is too risky at lunar return velocities. If you want to do aerobraking, you're back to Orion. Fundamentally speaking, the problem is that SpaceX hasn't built a moon optimized upper stage. Someone needs to build a crew transporter that is optimized for LEO -> NRHO -> LEO, like the cislunar transporter, but that one only carries fuel to NRHO.
  - devoutsalsa 2y ago
    >> Why are you so oddly fixated on SLS? Because I'm not a rocket scientist & if I knew how this stuff worked I wouldn't have to ask!
  - ikekkdcjkfke 2y ago
    It's insane that there were multiple crewed trips to the moon, in the 70's!
ChrisArchitect 2y ago
[dupe] More discussion: https://news.ycombinator.com/item?id=41370982 https://news.ycombinator.com/item?id=41370982
BurnGpuBurn 2y ago
I cannot fathom a world where this could happen without there being an enormous amount of fraud and corruption going on.
rdtsc 2y ago
> The report notes that NASA has declined to exercise an option to convert the contract to a fixed-price mechanism. They know besides the unrealistic high price they’d get shoved in their face by Bechtel, it would seriously jeopardize their chances of “retiring” at Bechtel with a comfortable salary later on.
demondemidi 2y ago
Just a reminder that no one will ever know how much private company SpaceX spends on their platforms because they are not required to divulge this info, nor how much their investors subsidize their government contracts (paid for by taxpayers).
advisedwang 2y ago
Bechtel were the contractor for Boston's Big Dig too. Why do people keep hiring these guys
- consumer451 2y ago
  https://www.opensecrets.org/federal-lobbying/clients/summary?id=D000000237 https://www.opensecrets.org/federal-lobbying/clients/summary...

45 points by NavinF 2y ago

>$2.7 billion. Such a cost is nearly twice the funding it took to build the largest structure in the world, the Burj Khalifa, which is seven times taller. > NASA officials informed us they do not intend to request a fixed-price proposal from Bechtel

gexla 2y ago
It seems a significant addition to the cost overruns is labor. Maybe this is a good example of the cost reality of bringing manufacturing back to the US. Given a difficult timeline, the US workers would need to be putting in a lot of overtime. Many of the laborers who took part in building the Burj Khalifa were being paid less than what the US workers would spend on a meal. https://en.wikipedia.org/wiki/Burj_Khalifa https://en.wikipedia.org/wiki/Burj_Khalifa Edit: Here is the NASA report. https://oig.nasa.gov/office-of-inspector-general-oig/audit-reports/nasas-management-of-the-mobile-launcher-2-project/ https://oig.nasa.gov/office-of-inspector-general-oig/audit-r...
- jiggawatts 2y ago
  > It seems a significant addition to the cost overruns is labor Only because they're billing for a ridiculous amount of time. Estimated completion time of 2027!? Wat? SpaceX has already built three and is planning more. By the time Bechtel completes their single tower, SpaceX will have half a dozen and a Moon colony. The contract should be torn up and the people that signed it barred from holding executive positions for life.
- omoikane 2y ago
  The article seems to be using the cost of Burj Khalifa without accounting for inflation. $1.5 billion in 2009 would be about $2.2 billion in 2024. NASA's project is still expensive, but not quite twice as expensive as Burj Khalifa (unless the budget overruns again).
  - disillusioned 2y ago
    The OIG report is projecting it could be as much as $2.7 Billion all said and done.

thot_experiment 2y ago

Someone who follows this stuff, how does this compare to the starship catching towers? Both in terms of complexity and cost. My gut says the starship towers are about as complicated (in an abstract "does a bunch of tricky things wrt spaceship support" even if they're different things way) and I'm betting it's 100x? 1000x cheaper? I don't know anything though so if someone is more knowledgeable I would love to learn more.

BHSPitMonkey 2y ago
SpaceX doesn't tend to share that kind of information, so we may never know exactly. Plus, it's R&D they expect will amortize over the operational life of Starship rather than infrastructure being built for a narrow set of missions that I can't see resulting in much future reuse.
glzone1 2y ago
SpaceX built (and rebuilt) a tower practically overnight. They are building two more. They are building huge construction bays overnight. Their contractors are literally working nights. There is probably no comparison on cost or speed. ML-1 and ML-2 have been percolating along for over a decade. ML-1 is going to see just a few launches (3-4?). ML-2 is just 7 feet taller. Total between the two maybe $3.5 billion range? This is just for the SLS tower. Actual SLS / Orion / sustainment costs at the NASA centers are absolutely mind boggling. Then you compare that to what NASA gets with their contracts with SpaceX and the mind boggles.
JumpCrisscross 2y ago
> how does this compare to the starship catching towers? These are mobile while Starship’s towers are fixed. That alone massively increases the MLs’ complexity. Why are they mobile? I genuinely can’t tell beyond that’s how Apollo did it. A fixed tower with a moveable (or even temporary) structure built around it would have been simpler.

JCharante 2y ago

Why do governments allow cost follow pricing? Companies should bid a final price and pay for insurance that covers unexpected costs/delays.

vimax 2y ago
How do you expect insurance companies to create actuarial tables for one-off government R&D projects?
- tedunangst 2y ago
  Insurance companies manage to find ways to price insurance for Hollywood actors storming off set.
- NavinF 2y ago
  Just estimate the probability of payout. Yes you can do this for non-repeating, hard-to-model events: https://www.astralcodexten.com/p/in-continued-defense-of-non-frequentist https://www.astralcodexten.com/p/in-continued-defense-of-non... I do it all the time without making actuarial tables, as do insurance companies.
- JCharante 2y ago
  Same way insurance companies insure satellite launches. Lots of guesses and getting another insurance company to underwrite their insurance deals.

ungreased0675 2y ago

NASA seems to be setting money on fire lately. It’s probably time to shutter the place and start over with the National Space and Air Administration.

readthenotes1 2y ago
About a decade ago and NASA employee worried to me that it was just becoming a white collar jobs program. Mission accomplished.

namlem 2y ago

SpaceX already built their launch/catch tower and didn't even bill the taxpayers for it.

glzone1 2y ago

Wild. But it's a pattern with these contracts. As a reminder, for SLS they decided to modify an exiting launch tower (ML-1). The original estimate to modify the tower was $54 million. Final cost will likely be $1 billion. And delayed of course. It's only going to get a few launches. You can see where this is going. Then they decided to do a new launch tower, ML-2. This tower is only 7 feet taller. They planned to incorporate lessons learned from the mess of ML-1 (haha). Started at $400M. Now going to be $2.7 billion (and delayed of course badly). These cost+ deals especially with multiple contractors are like rotting fruit that attracts maggots that glom on and provide negative value. Something that would take a day to a week elsewhere could be literally months. Every incentive is to move slowly. The contractor layers and paperwork needs on even basic changes are totally massive. The other thing that really jumps out is usually speed - these things will take FOREVER. I once even saw something about 8 track data recording in a govt contract, It's wild what is stacked into these things. The overhead to do anything can be wild. I once worked a govt job, and we needed an ipad for whatever reason (think something like foreflight on an ipad - most normal people would just buy an ipad to run foreflight). For all sorts of dumb and dumber reason, an ipad could not be purchased. I'd estimate $25K - $50K maybe went into trying to buy this damn ipad and coming up with workarounds. Can only go through central IT which wasn't even selected for price performance, has no customer service, and doesn't have apple SKU's. That's just the beginning. By the time you've wasted attorney time, contracting teams time on this ipad purchase and your own time, your managers time, all the downline folks time... the mind boggles. I think the final solution was to do a contract with a third party for a service, who could then buy the ipad and install the app on it as long as that part of the contract was less than some capital item threshold, and then that third party could enter into another agreement to make it available for their use. But the time to do this, then the time to come up with the agreement to use the ipad owned by this third party... total joke! I understood why teachers buy their own pencils. I'm sure a school can provide them, but a teacher probably doesn't want to deal with the headaches!

disillusioned 2y ago
Fire everyone at NASA involved in "project management" that allowed this to happen: >For the ML-2 project’s first 5 years, NASA lacked a reliable cost and schedule estimate, making it difficult for the Agency to accurately identify ML-2 budget requirements, be accountable to Congress and other stakeholders, and accurately measure project and contractor performance. Absolute dereliction of duty and complete incompetence. Bechtel has literally no _true_ incentive to deliver on time: the "awards" are folly and they've been awarded some of them even despite this abysmal performance. Instead, you can inflate the "cost" part of "cost plus" and just continue to suck on the teat of the taxpayer, while blaming supply chain and other challenges, all to not build what amounts to a several hundred foot tall elevator. (Obviously a gross simplification; I understand there are interconnected data, electrical, hydraulic, fluid, and other systems, along with being immediately adjacent an extremely high-thrust launch, but _still_.)

devoutsalsa 2y ago

If SLS is only getting a few launches, and we’re planning to establish a permanent presence on the moon, what will replace SLS? Starship and New Glenn?

imtringued 2y ago
Why are you so oddly fixated on SLS? SLS is a launch vehicle for Orion. What is needed is a replacement for Orion so NASA can ditch SLS. The only alternative to Orion is to use a capsule to LEO, transfer crew to a second fully fueled Starship, dock with HLS in NRHO, do the landing and come back, dock with the crew Starship and then do a deceleration burn, dock in LEO with the capsule and then return to earth. Why? Because the lunar Starship doesn't have enough delta V to return LEO and landing crew on earth via aerobraking is too risky at lunar return velocities. If you want to do aerobraking, you're back to Orion. Fundamentally speaking, the problem is that SpaceX hasn't built a moon optimized upper stage. Someone needs to build a crew transporter that is optimized for LEO -> NRHO -> LEO, like the cislunar transporter, but that one only carries fuel to NRHO.
- devoutsalsa 2y ago
  >> Why are you so oddly fixated on SLS? Because I'm not a rocket scientist & if I knew how this stuff worked I wouldn't have to ask!
- ikekkdcjkfke 2y ago
  It's insane that there were multiple crewed trips to the moon, in the 70's!

ChrisArchitect 2y ago

[dupe] More discussion: https://news.ycombinator.com/item?id=41370982 https://news.ycombinator.com/item?id=41370982

BurnGpuBurn 2y ago

I cannot fathom a world where this could happen without there being an enormous amount of fraud and corruption going on.

rdtsc 2y ago

> The report notes that NASA has declined to exercise an option to convert the contract to a fixed-price mechanism. They know besides the unrealistic high price they’d get shoved in their face by Bechtel, it would seriously jeopardize their chances of “retiring” at Bechtel with a comfortable salary later on.

demondemidi 2y ago

Just a reminder that no one will ever know how much private company SpaceX spends on their platforms because they are not required to divulge this info, nor how much their investors subsidize their government contracts (paid for by taxpayers).

advisedwang 2y ago

Bechtel were the contractor for Boston's Big Dig too. Why do people keep hiring these guys

consumer451 2y ago
https://www.opensecrets.org/federal-lobbying/clients/summary?id=D000000237 https://www.opensecrets.org/federal-lobbying/clients/summary...