• Sonori@beehaw.org
    link
    fedilink
    arrow-up
    7
    arrow-down
    1
    ·
    edit-2
    8 months ago

    Especially because in practice the massive cost of compute means that while low cost power is nice, it is always more economically sound from the miners prospective to run 24/7 to get the best return on investment. Indeed if you only ran your expensive data center for the few hours a day where solar production can even meet demand than you would never break even.

    All of this means that miners in practice have to demand all that power 24/7, and thusly demand either more storage or more fossil production than would otherwise be necessary for the grid. As such crypto mining is not incentivizing a move to renewables but forestalling it.

    If nothing else hydrogen is pitifully easy to produce from power, and current non-fossil hydrogen production is nowhere near enough to fulfill current demand, let alone what we’ll need in the next decades. Better yet, because the machinery needed to make hydrogen is so cheap compared to the electricity necessary, it actually makes financial sense to only run them when there is spare renewable capacity as compared to constantly.

    • jonne@infosec.pub
      link
      fedilink
      arrow-up
      2
      ·
      8 months ago

      Yeah, I’ve been wondering if instead of batteries homes could have their own electrolysis + fuel cell setup. Which one would be able to store more energy for a certain footprint?

      • Sonori@beehaw.org
        link
        fedilink
        arrow-up
        5
        ·
        8 months ago

        Probably not, while hydrogen production is simple enough by industrial standards, in practice a small hydrogen plant is still the size of a shipping container thanks to the need to reach cryogenic temperatures for decent density and like nearly all very complex industrial kit requires expert maintenance. It also has to deal with practical limitations on round trip efficiency and such that make it hard to scale down to the size where it would be useful for a home or small business.

        By contrast LFP or sodium ion batteries and inverters have no moving parts or seals to lubricate and replace or else the building explodes, no real maintenance needs, are far more energy dense, and offer more burst capacity than a fuel cell can.

        While it may be viable for grid scale storage, though given most practical proposals involve combustion turbines and get half the energy out you put in I’m skeptical they can compete with more efficient methods like pumped hydro, batteries, and long distance transmission, at least for home use I can’t see hydrogen being practical anytime soon.

        The real value in green hydrogen is for industrial uses like fertilizer production, ship fuel, and such, not in consumer applications.

    • vividspecter@lemm.ee
      link
      fedilink
      arrow-up
      2
      ·
      8 months ago

      Especially because in practice the massive cost of compute means that while low cost power is nice, it is always more economically sound from the miners prospective to run 24/7 to get the best return on investment.

      If you wanted a market solution, you could push wholesale power pricing onto miners so the cost potentially becomes too expensive to mine at times there isn’t renewable energy. Not that I think power should be wasted on crypto mining in any case.