8 ms·

Because it's a wrong assumption. Significant cost come from the assembly and testing of all the components itself, that you would have to redo entirely. Also i

Because it's a wrong assumption. Significant cost come from the assembly and testing of all the components itself, that you would have to redo entirely. Also if you were to build just 2 or 3 of them, you can't expect any economy of scale. On top of that, the operational cost of JWST is expected to be around 1B$ for it's lifetime, you could expect that to be similar for every single replica you have. And finally, you can only put one per rocket, and just the rocket is about 200M$ dollar, and you need to add all the cost of shipping the telescope to Guyana, that's not cheap. So overall, while a second replica would not cost you another 10B, it would probably cost in the order of 3-5B$, that's a lot of billions for a telescope with exactly identical capabilities to another one. It would still be useful, as astronomers are going to have to compete for time on the JWST and not everyone will be served, but the benefits of a second one would be marginal compared to the benefits of the first one. So the price/benefit ratio might actually be worse on a new copy. Meanwhile, there is a myriad of other very cool NASA projects that would greatly benefit from 3-5B$ instead and do things that we haven't done so far.

bumby 5y ago
This question comes up in every one of these threads and this is the correct response. If you look at the line items on a build, you might see something like a $200 bolt. It’s not that there was $198 of R&D going into the design of the bolt, it’s that quality management drives the cost. Chain-of-custody, bonding, material testing, witnessing etc. are all part of that effort and they don’t scale like a design spec does.
- WalterBright 5y ago
  If it's a one-off bolt with a custom design, it will likely cost thousands of dollars.
  - bumby 5y ago
    The GPs point was that subsequent builds would make it no longer a one-off design. My point is that there are other substantial cost drivers that break their assumptions. My analogy of a $200 bolt was not meant for a custom design, but the point stands regardless. (I’ve worked in a custom machine shop for aerospace, and depending on the tolerances, the actual build is typically not thousands until you factor in all the aspects in my previous post)
WalterBright 5y ago
Assuming your figures are correct, you'll get double the data for another 30% cost. If you wait a month before launching #2, if problems appear in #1 (like the telescope mirror was ground improperly) it can be fixed in #2. The operational cost will not double. The same ground facility, equipment, and staff can manage both. If you're buying two identical launches, you can get a quantity discount. > the benefits of a second one would be marginal compared to the benefits of the first one And yet I read many glowing accounts about how much extra value came from extending the Hubble's lifetime. I do have some experience with this. I worked for 3 years on the design of the 757. Thousands of engineers spending maybe 5 years on it. None of that has to be repeated. In a machine shop, most of the cost is in the setup. Making two adds little cost. I had a job assembling electronics to help pay for college. The first board would take 2 hours to build. The next one, half that. The fourth, 20 minutes. I'm sure there's plenty of software on board that machine. On HN we all know how expensive making software is. Making a copy costs nothing.
- lisper 5y ago
  > double the data Yes, but the data has diminishing returns as well. The real value in a second unit is having a backup in the event of a total failure of the first. But that's a hard sell to the taxpayers who fund these things.
  - WalterBright 5y ago
    > Yes, but the data has diminishing returns as well. I find that hard to believe considering there's literally a universe of things to look at. We find surprising things everywhere we look.
    - Beldin 5y ago
      $3-5B would go a long way to a new Einstein telescope (ie. gravitational waves detector). It would likey allow for significant sensitivity increases in the existing ones. It could propel the National Ignition Facility or ITER forward. It could be used to build a new Arecibo. It could add an enormous amount of low-frequency telescopes - or other scopes for very large integrated scope arrays. I'd much rather broaden our view than double it in one narrow band. (Or advance fusion research.)
BurningFrog 5y ago
> the benefits of a second one would be marginal compared to the benefits of the first one. This can't be true. The benefit of the JWST is the observations it can make Two JWSTs can make twice as many observations as one, so it provides twice the benefit. In some sense there is a diminishing return in that the most important observations will be attempted first, and and over time the average observation will be less and less important. But surely there is many decades of pent up very important research!
- sailingparrot 5y ago
  JWST was built to see in IR to help us get answers to a set of important questions. The best known one being trying to elucidate the question around the rate of expansion of the universe. We hope to get some answer by looking at objects much further away in space and time with the JWST and getting new estimates for the expansion rate. Something we cannot do today. You don't need two telescopes to do that, you just need one with a specific set of IR capabilities. Sure, having twice the imaging power is better, but it's definitely far from doubling the benefits. The lifetime of the JWST is expected to be 10+ years, that's a lot of data that will come to us already, and everything the astronomer community deems important will have time on the telescope. Just like we only needed 1 LHC to confirm the existence of the Higgs boson and the robustness of the standard model. Building two of those would have been a massive waste of money, it was much better to build one, run experiments, assess the results and then use the money that was saved by building a single one to build new tools with new capabilities to answer the new questions. Of course the reality of government budgeting is a little bit more complicated than my rosy picture but the point stands.
  - thiagotomei 5y ago
    A small nitpick: yes, we only needed 1 LHC to confirm the existence of the Higgs boson, BUT we made sure to have two experiments (ATLAS and CMS) looking for it. As far as I know, every modern high energy physics accelerator has had two (or more) sister experiments to cross-check each others! Of course, it is not a perfect analogy, since the two experiments are not replicas. They try to address the same physics cases, but they were designed, built and are operated in a completely independent way.
godelski 5y ago
Honestly that doesn't sound like much money. Especially since these projects aren't solely funded by one country, even if one foots most of the bill. Also, $1bn over it's lifetime is really cheap even just for the US alone.
- sailingparrot 5y ago
  > Honestly that doesn't sound like much money Well by comparison, here is the cost of two of the most impactful recent-ish (post 2000) space telescopes the US has launched: - Spitzer: 700M$, JWST being it's successor. This telescope allowed us to detect an exoplanet through light for the first time, refine our understanding of the shape of the milky way, find candidate objects to be further observed by JWST and many more contributions. - Kepler: 600M$, this is the telescope that allowed us to understand that planets were not rare at all, detecting more than 2500. So imagine what you can do with 3 to 5B$. Certainly more interesting things than just doubling your data gathering rate of a single telescope.
  - godelski 5y ago
    Instead of comparing to other telescope projects maybe better to compare to other mega science projects. Or better yet, other government costs. Though of course we'd need to run a cost benefit analysis. The thing is that these numbers are big for us and most companies, but these are government numbers and decades of work. Both factors are important. It's $11bn over 24 years, or 458m/yr. I'm happy to pay an extra $3/yr for this project.
    - sailingparrot 5y ago
      > I'm happy to pay an extra $3/yr for this project. Yes, I think many of us would be happy to. But the reality is that NASA has to fight a pretty ferocious budget battle every year, and often (not always) delay and increase in JWST cost has meant postponing or cancelling other missions. For example the Nancy Grace Roman Space Telescope, which was deemed the top priority by the decadal survey in 2010 almost got canceled in 2018/2020 because of JWST overruns. This Nature article is a good read: "The telescope that ate astronomy" [1] (and at the time of the article, JWST cost was "only" at 5B$). I am very excited about what we will observe with the JWST, but shelling yet a few other billion of dollars out of NASA's tight astronomy budget to get an identical copy, mean we are yet again cancelling or postponing other exciting missions that could help us answer very important questions. [1]: https://www.nature.com/articles/4671028a https://www.nature.com/articles/4671028a
WalterBright 5y ago
I remember a NOVA on one of the Mars landers. Nobody had ever landed anything on Mars with a parachute before, so the developers had a massive problem. They built this humongous facility to test parachutes. One design after another failed, for several months. They finally found a design that worked. I don't know what building a second chute would cost, but I bet it would be less than one thousandth of the cost of #1.

bumby 5y ago

This question comes up in every one of these threads and this is the correct response. If you look at the line items on a build, you might see something like a $200 bolt. It’s not that there was $198 of R&D going into the design of the bolt, it’s that quality management drives the cost. Chain-of-custody, bonding, material testing, witnessing etc. are all part of that effort and they don’t scale like a design spec does.

WalterBright 5y ago
If it's a one-off bolt with a custom design, it will likely cost thousands of dollars.
- bumby 5y ago
  The GPs point was that subsequent builds would make it no longer a one-off design. My point is that there are other substantial cost drivers that break their assumptions. My analogy of a $200 bolt was not meant for a custom design, but the point stands regardless. (I’ve worked in a custom machine shop for aerospace, and depending on the tolerances, the actual build is typically not thousands until you factor in all the aspects in my previous post)

WalterBright 5y ago

Assuming your figures are correct, you'll get double the data for another 30% cost. If you wait a month before launching #2, if problems appear in #1 (like the telescope mirror was ground improperly) it can be fixed in #2. The operational cost will not double. The same ground facility, equipment, and staff can manage both. If you're buying two identical launches, you can get a quantity discount. > the benefits of a second one would be marginal compared to the benefits of the first one And yet I read many glowing accounts about how much extra value came from extending the Hubble's lifetime. I do have some experience with this. I worked for 3 years on the design of the 757. Thousands of engineers spending maybe 5 years on it. None of that has to be repeated. In a machine shop, most of the cost is in the setup. Making two adds little cost. I had a job assembling electronics to help pay for college. The first board would take 2 hours to build. The next one, half that. The fourth, 20 minutes. I'm sure there's plenty of software on board that machine. On HN we all know how expensive making software is. Making a copy costs nothing.

lisper 5y ago
> double the data Yes, but the data has diminishing returns as well. The real value in a second unit is having a backup in the event of a total failure of the first. But that's a hard sell to the taxpayers who fund these things.
- WalterBright 5y ago
  > Yes, but the data has diminishing returns as well. I find that hard to believe considering there's literally a universe of things to look at. We find surprising things everywhere we look.
  - Beldin 5y ago
    $3-5B would go a long way to a new Einstein telescope (ie. gravitational waves detector). It would likey allow for significant sensitivity increases in the existing ones. It could propel the National Ignition Facility or ITER forward. It could be used to build a new Arecibo. It could add an enormous amount of low-frequency telescopes - or other scopes for very large integrated scope arrays. I'd much rather broaden our view than double it in one narrow band. (Or advance fusion research.)

BurningFrog 5y ago

> the benefits of a second one would be marginal compared to the benefits of the first one. This can't be true. The benefit of the JWST is the observations it can make Two JWSTs can make twice as many observations as one, so it provides twice the benefit. In some sense there is a diminishing return in that the most important observations will be attempted first, and and over time the average observation will be less and less important. But surely there is many decades of pent up very important research!

sailingparrot 5y ago
JWST was built to see in IR to help us get answers to a set of important questions. The best known one being trying to elucidate the question around the rate of expansion of the universe. We hope to get some answer by looking at objects much further away in space and time with the JWST and getting new estimates for the expansion rate. Something we cannot do today. You don't need two telescopes to do that, you just need one with a specific set of IR capabilities. Sure, having twice the imaging power is better, but it's definitely far from doubling the benefits. The lifetime of the JWST is expected to be 10+ years, that's a lot of data that will come to us already, and everything the astronomer community deems important will have time on the telescope. Just like we only needed 1 LHC to confirm the existence of the Higgs boson and the robustness of the standard model. Building two of those would have been a massive waste of money, it was much better to build one, run experiments, assess the results and then use the money that was saved by building a single one to build new tools with new capabilities to answer the new questions. Of course the reality of government budgeting is a little bit more complicated than my rosy picture but the point stands.
- thiagotomei 5y ago
  A small nitpick: yes, we only needed 1 LHC to confirm the existence of the Higgs boson, BUT we made sure to have two experiments (ATLAS and CMS) looking for it. As far as I know, every modern high energy physics accelerator has had two (or more) sister experiments to cross-check each others! Of course, it is not a perfect analogy, since the two experiments are not replicas. They try to address the same physics cases, but they were designed, built and are operated in a completely independent way.

godelski 5y ago

Honestly that doesn't sound like much money. Especially since these projects aren't solely funded by one country, even if one foots most of the bill. Also, $1bn over it's lifetime is really cheap even just for the US alone.

sailingparrot 5y ago
> Honestly that doesn't sound like much money Well by comparison, here is the cost of two of the most impactful recent-ish (post 2000) space telescopes the US has launched: - Spitzer: 700M$, JWST being it's successor. This telescope allowed us to detect an exoplanet through light for the first time, refine our understanding of the shape of the milky way, find candidate objects to be further observed by JWST and many more contributions. - Kepler: 600M$, this is the telescope that allowed us to understand that planets were not rare at all, detecting more than 2500. So imagine what you can do with 3 to 5B$. Certainly more interesting things than just doubling your data gathering rate of a single telescope.
- godelski 5y ago
  Instead of comparing to other telescope projects maybe better to compare to other mega science projects. Or better yet, other government costs. Though of course we'd need to run a cost benefit analysis. The thing is that these numbers are big for us and most companies, but these are government numbers and decades of work. Both factors are important. It's $11bn over 24 years, or 458m/yr. I'm happy to pay an extra $3/yr for this project.
  - sailingparrot 5y ago
    > I'm happy to pay an extra $3/yr for this project. Yes, I think many of us would be happy to. But the reality is that NASA has to fight a pretty ferocious budget battle every year, and often (not always) delay and increase in JWST cost has meant postponing or cancelling other missions. For example the Nancy Grace Roman Space Telescope, which was deemed the top priority by the decadal survey in 2010 almost got canceled in 2018/2020 because of JWST overruns. This Nature article is a good read: "The telescope that ate astronomy" [1] (and at the time of the article, JWST cost was "only" at 5B$). I am very excited about what we will observe with the JWST, but shelling yet a few other billion of dollars out of NASA's tight astronomy budget to get an identical copy, mean we are yet again cancelling or postponing other exciting missions that could help us answer very important questions. [1]: https://www.nature.com/articles/4671028a https://www.nature.com/articles/4671028a

WalterBright 5y ago

I remember a NOVA on one of the Mars landers. Nobody had ever landed anything on Mars with a parachute before, so the developers had a massive problem. They built this humongous facility to test parachutes. One design after another failed, for several months. They finally found a design that worked. I don't know what building a second chute would cost, but I bet it would be less than one thousandth of the cost of #1.