10 ms·

This is a classic example of sub-optimization (and even if it's hand-wavey pseudo-science and may be wrong, the general concept is still interesting). Each ind

This is a classic example of sub-optimization (and even if it's hand-wavey pseudo-science and may be wrong, the general concept is still interesting). Each individual wind turbine is optimized to be the "most efficient" it can possibly be. But not in the context of the environment, which requires complex control systems and methods to reduce damage in non-optimal conditions. And not if you take cost and efficiency of the whole system into account; where four times as many small turbines with less complexity run more often and produce more overall output. But hey, each wind turbine is "optimal." Interesting. Most complex systems have this property. Even (especially?) your business.

nkoren 10y ago
This kind of thinking contributed to American rocketry being trapped in a local maxima for decades. There are many reasons it became trapped, but one of the primary reasons was that rocket scientists would insist on optimising rocket engines at the expense of every other aspect of the vehicle. "The logic of rocket equation is brutal and inexorable", the NASA rocket scientists would say. "As the performance of the rocket engine decreases linearly, the propellent requirements increases exponentially. Therefore it is imperative to have the highest-possible performance from your rocket engines. The combustion of liquid hydrogen / oxygen is the highest-efficiency chemical reaction, and therefore liquid hydrogen is the only propellant we will consider using for our engines." All of this was correct, but also wrong. It was wrong in a whole-systems context, because: A.) Liquid hydrogen is much lower-density than traditional rocket fuels like Kerosene, requiring far larger and heavier structures to carry the same amount of energy, and larger and heavier engines and plumbing and everything else -- and the structural mass of a rocket matters. B.) Liquid hydrogen will boil off and leak through anything, requiring massive amounts of insulation (again, adding weight), and far stricter operational protocols around the fueled vehicle. C.) Liquid hydrogen is so cold that it will embrittle and shatter ordinary metals and requires much more exotic and expensive metallurgy. and finally, D.) The alternative, kerosene, gives you less efficient engines but also much lighter-weight vehicles -- and sure, per that inexorable logic of the rocket equation, you need to use more fuel for the same amount of payload, but kerosene is cheaper than milk so who fucking cares? Anyhow, the upshot of this is that US rocket scientists spent literally tens of billions of dollars developing hyper-locally-optimised rocketry schemes, most of which failed outright. For the remainder, for every dollar of kerosene they didn't have to buy, they probably spent upwards of $1000 on exotic aerospace hardware. Finally, they gave up and just bought kerosene-based engines from the Russians. So then there's Elon Musk. And of course there's a ton of brilliant rocket science in the Falcon, but much of the reason for SpaceX's success is because it doesn't optimise solely on the rocket science. They optimise for cost as a whole, and consider factors like design, manufacturing, operations, reusability, etc. to be relevant to the question of cost. For example: the Falcon 9 would undoubtedly be more efficient if it had a hydrogen-based upper stage. But doing that would cause it to lose most of its commonality with the lower stage, requiring an entirely new design, manufacturing, and operations workforce. Which would cost several orders of magnitude more than the extra kerosene required to fly a "sub-optimal" rocket. The reason they've already cut the price of launch by about 80% vs. (say) the Space Shuttle is because they optimised the whole system, keeping the unique part count and operational complexity as low as possible. I still sometimes hear old-school rocket scientists grumble about how the Falcon is a less optimised vehicle than the Shuttle, though. They just don't get it. Never will. Anyhow, that was a tangent. Fascinating to think that wind turbines might be amenable to a similar class of disruption.
- masklinn 10y ago
  > Anyhow, that was a tangent It's also untrue, US rocketry kept varied propulsion types throughout its history including both RP1-only and LH2-only as well as mixed stages, HTPB and Aerozine. Your painting of rocket scientists as some sort of confederacy of dunce unable to get a grip on tradeoffs is not only inane it's insulting bullshit.
  - HCIdivision17 10y ago
    I started reading Ignition! and it is a hoot. I really get the impression that the rocket fuel scientists fixated, not because of personal issues (though hilarity ensues there), but because the search space is damn large (at least while still discovering how fuels interacted). Or worse, it just takes a damn long time to even properly test. So when you found something promising, you really dug into it.
    - sevensor 10y ago
      I loved the section on mercaptans -- they persisted way longer than they needed to on experiments with some of the worst smelling chemicals anywhere.
    - nkurz 10y ago
      Yes, it's a fantastic book. Not necessarily a good role model for future work, but it perfectly captures the spirit of rocket science at the time: http://library.sciencemadness.org/library/books/ignition.pdf http://library.sciencemadness.org/library/books/ignition.pdf Everyone who grew up idolizing rocket scientists should read it. If you have problems with missing pages or broken figures, try a different PDF viewer. The PDF has something odd about it that confuses some in-browser viewers.
  - mistermann 10y ago
    I'm confused now....so you're saying that in the history of US rocket implementations (in actual deployed projects), the optimal design was used every time (and therefore the comment you are replying to is completely incorrect)?
hencq 10y ago
Yeah, another example that comes to mind is Combined Heat and Power (CHP) plants, where the efficiency of power generation is actually lower, but the heat that's normally wasted is made useful.

nkoren 10y ago

This kind of thinking contributed to American rocketry being trapped in a local maxima for decades. There are many reasons it became trapped, but one of the primary reasons was that rocket scientists would insist on optimising rocket engines at the expense of every other aspect of the vehicle. "The logic of rocket equation is brutal and inexorable", the NASA rocket scientists would say. "As the performance of the rocket engine decreases linearly, the propellent requirements increases exponentially. Therefore it is imperative to have the highest-possible performance from your rocket engines. The combustion of liquid hydrogen / oxygen is the highest-efficiency chemical reaction, and therefore liquid hydrogen is the only propellant we will consider using for our engines." All of this was correct, but also wrong. It was wrong in a whole-systems context, because: A.) Liquid hydrogen is much lower-density than traditional rocket fuels like Kerosene, requiring far larger and heavier structures to carry the same amount of energy, and larger and heavier engines and plumbing and everything else -- and the structural mass of a rocket matters. B.) Liquid hydrogen will boil off and leak through anything, requiring massive amounts of insulation (again, adding weight), and far stricter operational protocols around the fueled vehicle. C.) Liquid hydrogen is so cold that it will embrittle and shatter ordinary metals and requires much more exotic and expensive metallurgy. and finally, D.) The alternative, kerosene, gives you less efficient engines but also much lighter-weight vehicles -- and sure, per that inexorable logic of the rocket equation, you need to use more fuel for the same amount of payload, but kerosene is cheaper than milk so who fucking cares? Anyhow, the upshot of this is that US rocket scientists spent literally tens of billions of dollars developing hyper-locally-optimised rocketry schemes, most of which failed outright. For the remainder, for every dollar of kerosene they didn't have to buy, they probably spent upwards of $1000 on exotic aerospace hardware. Finally, they gave up and just bought kerosene-based engines from the Russians. So then there's Elon Musk. And of course there's a ton of brilliant rocket science in the Falcon, but much of the reason for SpaceX's success is because it doesn't optimise solely on the rocket science. They optimise for cost as a whole, and consider factors like design, manufacturing, operations, reusability, etc. to be relevant to the question of cost. For example: the Falcon 9 would undoubtedly be more efficient if it had a hydrogen-based upper stage. But doing that would cause it to lose most of its commonality with the lower stage, requiring an entirely new design, manufacturing, and operations workforce. Which would cost several orders of magnitude more than the extra kerosene required to fly a "sub-optimal" rocket. The reason they've already cut the price of launch by about 80% vs. (say) the Space Shuttle is because they optimised the whole system, keeping the unique part count and operational complexity as low as possible. I still sometimes hear old-school rocket scientists grumble about how the Falcon is a less optimised vehicle than the Shuttle, though. They just don't get it. Never will. Anyhow, that was a tangent. Fascinating to think that wind turbines might be amenable to a similar class of disruption.

masklinn 10y ago
> Anyhow, that was a tangent It's also untrue, US rocketry kept varied propulsion types throughout its history including both RP1-only and LH2-only as well as mixed stages, HTPB and Aerozine. Your painting of rocket scientists as some sort of confederacy of dunce unable to get a grip on tradeoffs is not only inane it's insulting bullshit.
- HCIdivision17 10y ago
  I started reading Ignition! and it is a hoot. I really get the impression that the rocket fuel scientists fixated, not because of personal issues (though hilarity ensues there), but because the search space is damn large (at least while still discovering how fuels interacted). Or worse, it just takes a damn long time to even properly test. So when you found something promising, you really dug into it.
  - sevensor 10y ago
    I loved the section on mercaptans -- they persisted way longer than they needed to on experiments with some of the worst smelling chemicals anywhere.
  - nkurz 10y ago
    Yes, it's a fantastic book. Not necessarily a good role model for future work, but it perfectly captures the spirit of rocket science at the time: http://library.sciencemadness.org/library/books/ignition.pdf http://library.sciencemadness.org/library/books/ignition.pdf Everyone who grew up idolizing rocket scientists should read it. If you have problems with missing pages or broken figures, try a different PDF viewer. The PDF has something odd about it that confuses some in-browser viewers.
- mistermann 10y ago
  I'm confused now....so you're saying that in the history of US rocket implementations (in actual deployed projects), the optimal design was used every time (and therefore the comment you are replying to is completely incorrect)?

hencq 10y ago

Yeah, another example that comes to mind is Combined Heat and Power (CHP) plants, where the efficiency of power generation is actually lower, but the heat that's normally wasted is made useful.