5 ms·

Your "repeated" refutations include a claim of $1400 per kWh-year of hydrogen storage with a link to a report [1]. That seemed way too high for me so I grabbed

Your "repeated" refutations include a claim of $1400 per kWh-year of hydrogen storage with a link to a report [1]. That seemed way too high for me so I grabbed the report - which, by the way, argues that hydrogen is economical for storage of spilled wind - and have no idea where you got this figure from. It appears you've mixed up kW and kWh at the very least. The figure you're reading also appears to include the cost of buying the electricity at market rates, which is shown to not be economical. But it isn't what parent argued for, and you explicitly state it does not include. Reading further in to the report, it breaks down the capital cost of a hydrogen system minus storage at $550/kW, and notes the very low cost of underground storage at large scales (for smaller scales, it argues for using tanks and being geographically independent). So you disagreed with parent, threw out a nonsensical figure and linked to a report as proof that agrees with them, not you. And hey, it looks like someone informed you of your mistake 3 months ago [2], which you didn't bother to reply to or correct yourself. This is a repeated theme with your accounts (your main and your sockpuppet). Every time I dig in to the numbers you throw out I get a sour taste in my mouth that you've caused me to waste my time. I believe that you are arguing in bad faith, cherry picking whatever numbers you can find, in context or not, to make a pro nuclear case. Which is a bit sad, because I think there are still strong (but dwindling) cases to be made for nuclear. [1] https://prod-ng.sandia.gov/techlib-noauth/access-control.cgi/2011/114845.pdf https://prod-ng.sandia.gov/techlib-noauth/access-control.cgi... [2] https://news.ycombinator.com/item?id=23741288 https://news.ycombinator.com/item?id=23741288

Manfredo_1 6y ago
Capital cost is the initial cost to build this system. The figure for comes from figure 14. The cost estimates are $3,000 with current technology. This works out to $500 per KWh 20 years system. To match the lifespan of a nuclear plant (typically 60 to 80 years), this works out to $1,500 to $2,000 per KWh when factoring the limited life span of the storage. Regarding the reply with no response, timestamps get more coarse grained the further in the past you get. This reply could easily be weeks after my initial comment, and I would have explained how I arrived at these figures if I has read it. Meanwhile, glaring inaccuracies like calming that existing natural gas infrastructure can be used to transport hydrogen go unmentioned. Or that "hardly anyone is building nuclear" when in fact most of the world by population is. And if we're talking about cherry-picking pfdietz incessantly cherry picks the Vogtle and Flamanville plants as though they're representative of historic nuclear costs. You arrive at the conclusion that I'm arguing in bad faith, while conveniently ignoring these issues with the claims I'm arguing against. This exchange makes me sad, too. Solar and wind serve as useful temporary mitigation of carbon output, even though they don't pave a way to full decarbonization. A hydrogen economy has a place to fulfill demand where energy density is necessary, like in transportation. But trying to advocate pipe-dream projects like storing tens of terawatt hours worth of energy in hydrogen really makes me wonder whether solar and wind advocates really have a grasp on the engineering challenges they're facing. Especially when these plans rest on outright falsehoods, like the claim that we can using existing natural gas pipes for hydrogen.
- reitzensteinm 6y ago
  Right, that's the wrong figure to use in the context of the linked debate. It's the cost to operate the plant, including buying power at market rates. The "spilled wind" costs are the cost of the plant itself used in the manner described, which equals $1400/kW - actually, this is probably where you got the number originally - or $233/kWh. However, this is over the lifetime of the plant, meaning the storage is closer to $12/kWh per year. That's not exactly cheap. You have to overbuild renewables to have the extra energy to store in it, and it's probably only going to cycle a few dozen times a year. Plus the $12/kWh figure likely represents (the study does not specify) the energy of the gas before conversion back to electricity which is lossy. But adding extra storage is cheap compared to adding extra power, so unlike with lithium ion batteries you can actually feasibly build weeks worth of storage in to the system. So, I'm not trying to say the numbers are great. I'm not saying this is what we should do. But there's a vast gap between the $1400 you quoted and the $12 figure which - at least I believe - is the correct number to use for this discussion. There are absolutely tons of people (especially on HN) that trivialize the problems around rolling out renewables. Probably the majority of my posts on the topic have actually been on the other side, pointing out issues or complaining about subsidies instead of carbon taxes. But I really think you've gone 180 degrees in the other direction and the numbers you use make it seem an order of magnitude harder than it is. I think even more extreme than using the worst Western nuclear cost overruns as representative of the intrinsic cost of nuclear.
  - pfdietz 6y ago
    For other cost figures for hydrogen storage, see the options and linked references at https://model.energy/ https://model.energy/ You can also use that site (and its weather time series) and play with the assumptions to see when nuclear can be made competitive with renewables + storage at supplying steady power output in various places, under a set of simplifying assumptions that tend to favor nuclear. You will find that hydrogen tends to be more important in high latitude places using wind.
  - Manfredo_1 6y ago
    $12 per kilowatt hour, per year. You then need to multiply by the number of years that this storage system needs to be online. Again, if we're comparing against the alternative of building a nuclear plant, this is 60 to 80 years. Neglecting to factor in the required lifespan artificially deflates the price drastically. And again this figure of $12 per KWh per year is for the "spilled wind" scenario - that is, the assumption that the electricity input is provided at zero cost. The other poster also did more than just cherry pick particularly expensive plants. Among other things, they claimed that existing natural gas infrastructure can be used to transport hydrogen. This is simply untrue, hydrogen would corrode metal pipes used for natural gas. There's a factor of 2,000 less hydrogen pipelines. Probably even less overall infrastructure if you count natural gas transported by rail and LNG shipping.

Manfredo_1 6y ago

Capital cost is the initial cost to build this system. The figure for comes from figure 14. The cost estimates are $3,000 with current technology. This works out to $500 per KWh 20 years system. To match the lifespan of a nuclear plant (typically 60 to 80 years), this works out to $1,500 to $2,000 per KWh when factoring the limited life span of the storage. Regarding the reply with no response, timestamps get more coarse grained the further in the past you get. This reply could easily be weeks after my initial comment, and I would have explained how I arrived at these figures if I has read it. Meanwhile, glaring inaccuracies like calming that existing natural gas infrastructure can be used to transport hydrogen go unmentioned. Or that "hardly anyone is building nuclear" when in fact most of the world by population is. And if we're talking about cherry-picking pfdietz incessantly cherry picks the Vogtle and Flamanville plants as though they're representative of historic nuclear costs. You arrive at the conclusion that I'm arguing in bad faith, while conveniently ignoring these issues with the claims I'm arguing against. This exchange makes me sad, too. Solar and wind serve as useful temporary mitigation of carbon output, even though they don't pave a way to full decarbonization. A hydrogen economy has a place to fulfill demand where energy density is necessary, like in transportation. But trying to advocate pipe-dream projects like storing tens of terawatt hours worth of energy in hydrogen really makes me wonder whether solar and wind advocates really have a grasp on the engineering challenges they're facing. Especially when these plans rest on outright falsehoods, like the claim that we can using existing natural gas pipes for hydrogen.

reitzensteinm 6y ago
Right, that's the wrong figure to use in the context of the linked debate. It's the cost to operate the plant, including buying power at market rates. The "spilled wind" costs are the cost of the plant itself used in the manner described, which equals $1400/kW - actually, this is probably where you got the number originally - or $233/kWh. However, this is over the lifetime of the plant, meaning the storage is closer to $12/kWh per year. That's not exactly cheap. You have to overbuild renewables to have the extra energy to store in it, and it's probably only going to cycle a few dozen times a year. Plus the $12/kWh figure likely represents (the study does not specify) the energy of the gas before conversion back to electricity which is lossy. But adding extra storage is cheap compared to adding extra power, so unlike with lithium ion batteries you can actually feasibly build weeks worth of storage in to the system. So, I'm not trying to say the numbers are great. I'm not saying this is what we should do. But there's a vast gap between the $1400 you quoted and the $12 figure which - at least I believe - is the correct number to use for this discussion. There are absolutely tons of people (especially on HN) that trivialize the problems around rolling out renewables. Probably the majority of my posts on the topic have actually been on the other side, pointing out issues or complaining about subsidies instead of carbon taxes. But I really think you've gone 180 degrees in the other direction and the numbers you use make it seem an order of magnitude harder than it is. I think even more extreme than using the worst Western nuclear cost overruns as representative of the intrinsic cost of nuclear.
- pfdietz 6y ago
  For other cost figures for hydrogen storage, see the options and linked references at https://model.energy/ https://model.energy/ You can also use that site (and its weather time series) and play with the assumptions to see when nuclear can be made competitive with renewables + storage at supplying steady power output in various places, under a set of simplifying assumptions that tend to favor nuclear. You will find that hydrogen tends to be more important in high latitude places using wind.
- Manfredo_1 6y ago
  $12 per kilowatt hour, per year. You then need to multiply by the number of years that this storage system needs to be online. Again, if we're comparing against the alternative of building a nuclear plant, this is 60 to 80 years. Neglecting to factor in the required lifespan artificially deflates the price drastically. And again this figure of $12 per KWh per year is for the "spilled wind" scenario - that is, the assumption that the electricity input is provided at zero cost. The other poster also did more than just cherry pick particularly expensive plants. Among other things, they claimed that existing natural gas infrastructure can be used to transport hydrogen. This is simply untrue, hydrogen would corrode metal pipes used for natural gas. There's a factor of 2,000 less hydrogen pipelines. Probably even less overall infrastructure if you count natural gas transported by rail and LNG shipping.