Hacker News Reader

Hacker News

Comment by ajsnigrutin | Hacker News Reader

Parent Full thread

ajsnigrutin

"but those take 10 years to build, that's too long"

-people telling me this 5 years ago

-also people telling me the same 10 years ago

-also people telling me the same 15 years ago

-also people telling me te same 20 years ago

-and of course, people telling me this now.

View on HN

Just like software developers, politicians focus too much on the short term.

From the article: “In a crisis, you have to decide appropriate to the situation and based on clear facts,” - see? in a crisis only.

You mean like in a global pandemic? (cries in fact based decision making)

That's democracy for you.

All political decisions in western democracies have a hard limit of a four year time horizon. Sure, there are times when collective decisions can be made with decades of planning - but only if they can fall through the increasingly narrow gaps between left/right politics.

Don't get me wrong, it's still better than any alternative I've seen anywhere, but it has its weaknesses.

> All political decisions in western democracies have a hard limit of a four year time horizon.

5 years in the UK, though it's been a long time since there was a change in power between parties without a 10+ year stint in office first.

Yeah, but that limits decision making to about 4 years if you have 5 year election cycles.

The last year you don’t want to piss off the voters so you have to be very sparse in getting anything done, so the last year in the election run up is spent pandering to as many people as possible rather than actually doing any of the real work.

Even if in the end you get multiple terms in office, you were never guaranteed re-election, so you were forced to make short term plans regardless (to stop the other lot getting in instead).

Merkel was in office for 16 years. Schröder 8 years. Kohl 16 years.

Correct?

Not sure I follow your point.

Elections happen every X years whether you win them or not. The job of a politician is to pander to the electorate, not take action.

Nobody cared about climate change 20 years ago, at least not enough to build nuclear power. People laughed at Al Gore and called it a hoax or denies it.

People only really started abandoning nuclear power with the advent of fukushima.

"Nobody" cared, yet the Vice President on the US was picked partly because of his strong credentials on the issue. Gore published "Earth in Balance" exactly 20 years ago, and even then it contained nothing new or controversial outside the Fox News bubble.

Nobody cares enough about climate change even now to actually do anything serious about it. Even in countries where both sides of politics pay lip service to agreeing for the need for radical changes they haven't been happening. And in others we still have entire political parties arguing against even minimal adjustments to our economies. We're basically hanging all our hopes on some sort of technological miracle.

People also tell to build wind and solar, while ignoring that those two needs to be backed up by coal and gas.

Does it? I feel like if we overbuild wind by like a factor of x5 in whole of Europe (on top of the median load factor, not assuming 100% of course), we would pretty much not need coal or gas. x5 ratio is still lower than nuclear reactor budget and time ratios overrun in the last 2 decades.

That being said I'm all for not putting all our eggs in the same basket, and having both nuclear, wind, solar AND coal and gas plants is fine by me (but the plan must say that the coal and gas plants will never be in use, but they are ready "just in case")

5x overbuild won't be enough, as I mentioned elsewhere at least in TX its not unusual for the wind to be producing at less than 15% of rated capacity. Particularly in the winter/summer when the energy load is at its highest.

So, like the 8% production two weeks ago, you need a 10x overbuild _just_ to reach rated capacity consistently. And you probably need some kind of long term storage as well for when it really goes bad.

There is a huge difference between slow to dispatch (coal/nuke/etc) plants which take hours to come up to full pressure, and simply not having anything that can come online (solar/wind) for a predicted weather event.

> 5x overbuild won't be enough, as I mentioned elsewhere at least in TX its not unusual for the wind to be producing at less than 15% of rated capacity. Particularly in the winter/summer when the energy load is at its highest.

Unless TX has a median load factor of 100% (in which case, wow), 5x overbuilding on top of median load factor definitely works in the case you describe.

As far as I know, the typical wind average load factor is 33%. So over-scaling by 5x of that would mean that 7% is still enough to provide. Granted, that's not a very large margin to the 8% you mention, but Europe is bigger than Texas, so it's less likely to have such a low-wind event.

Your just restating (or not understanding) what I said, the average is misleading because its seasonal, which isn't even clear unless you look for it in the graphs (because they swing). edit: (and I should be clear, its still swinging when the daily average is 5%, it might say produce 8% during the evening but only 2% during the day).

Here look at whats going on:

https://www.eia.gov/electricity/gridmonitor/expanded-view/cu...

You will have to tweak the graph because I didn't spend the time to figure out how to create a link to a graph which isolates just wind, and then expands the time-frame out to an entire year.

If you do that, what you see is during the summer the lulls are much lower and last for much longer. AKA it might go for a week or two basically not producing anything. It happens at other times of the year too, but during the spring/fall the weather is changing faster (aka its less stable) and the running average goes up.

This is what the wind/solar people seem to refuse to understand, its not a simple averaged calculation over a year because the worse case is much much worse for storage and capacity. The wind blows at 5% nameplate for two weeks, and the amount of storage and overbuild required to bridge that makes it an order of magnitude more expensive than pretty much every other source. Made massively worse because in say TX, that is also when energy consumption peaks.

There is median 29 hours of sun during December in Poland. The daily amplitude means nothing here.

Yet we build 10GW of capacity, that mostly does not contribute towards lower coal usage.

And, wind capacity factor is usually very low.

You'll notice that I mentioned over-scaling wind, not solar. I'm not a fan of solar overall, except maybe some clever big projects like using deserts, and for personal self-sufficiency (where people would over-scale a lot).

I don't mind keeping building solar panels, but subsidizing requires proper analysis of the benefits and externalities (notably on the grid) rather than blindly giving money to anyone saying !SOLAR! like we currently do.

Solar in deserts will shortly be recognized as a very dumb idea.

Solar on reservoirs and in pastures will be recognized as an overwhelmingly better idea.

A dollar spent on solar displaces several times as much carbon as anything else except wind, even after accounting for night and weather.

Wind capacity regularly drops below 10% in Poland. https://app.electricitymaps.com/zone/PL

You only need to overbuild the total energy requirement + the inefficiency cost of H2. Then you generate + store H2.

What do we do when there is no wind?

If there's a continent scale lack of wind in Europe I would be more afraid about the general state of our sun.

Try find a windless day here.

https://www.windy.com/?50.345,6.328,3

You mean like half of 2021? https://theconversation.com/what-europes-exceptionally-low-w...

Thanks for the article, I didn't know climate models estimated EU will have less wind, I expected we would get more wind.

That being said, the article is pretty low on numbers. The only actual number shown is -32% in UK. Which can totally be overcome by having 5x overproduction in whole of Europe.

We of course need to overbuild. You can do x4 for off shore wind and still come out ahead of nuclear. That map also corresponds to the time of year solar works 16 hours per day in Europe. It also shows that France Italy and the eastern part of the Baltics had more wind than usual.

Wind capacity regularly drops below 10% in Poland. https://app.electricitymaps.com/zone/PL

Poland is about the worst country in the EU for on-shore wind so a very cherrypicked counter example.

Do you have numbers for off-shore parks along the Baltic coast? That would be much more interesting.

Again... what do you do, when there's no wind along the baltic coast? Even a single windless night? Do you turn peoples power off, until it starts blowing again?

Wind on the north sea? Or northern Baltic? Or Swedish or Norwegian hydro? Or alps based hydro?

We are talking about a continent scale grid.

If we'll be fully dependant on other countries for our energy this will never pass. Energy independency is a must.

So I presume you are for Poland to exit the EU also? I am not talking about relying on Russia here.

Why, does it forbid energy independency somehow?

And what's wrong with idea that we shouldn't freely give other countries leverage on us?

> in whole of Europe

was presumably the answer to that. Now of course one does need a proper analysis of weather patterns (wind+sun) over the whole of Europe over the long term to know how well it would actually work. But it is a fairly large area.

In a world of increasingly broken supply chains, insane inflation, and gross materials shortages, I would never be hypothesizing about over-building anything.

or batteries / battery analogs. It is possible to solve this problem with tools we currently have. If we're thinking long term, we do not require coal or gas for power generation given sufficient hydro, wind, solar and storage. Cryogenic air, pumped hydro, and other storage designs are possible.

Chemical batteries are completely useless and expensive gimmick in state-wide storage of electric energy. Cost is main reason why it is cheaper to build coal/gas plant instead.

From what I've read batteries are the preferred option for short-term zero-emissions levelling but once you get into longer term stuff you start looking at storing compressed air or hydrogen in salt caverns. But hydrogen cracking/storage as a grid-level utility is still at the experimental stage.

Power-to-gas (Hydrogen or Methane) as battery is from my perspective only viable solution. Everything else appears to be more or less elaborate attempt to suck subsidies, but provide inadequate system.

Here is a cool pipe dream: https://heindl-energy.com/

The only gravity storage that isn't nonsense is pumped hydro.

https://www.youtube.com/watch?v=iGGOjD_OtAM

Right when you hear about power problem in TX, its _always_ when the wind isn't blowing. All that wind power has slowed investment in anything that isn't a gas plant, yet its all those gas plants running at 100% capacity and the wind basically producing at <10% which are a problem. Because apparently ERCOT is just adding the wind power (well based on some statistical weather analysis) to the total state capacity instead of assuming its 0% in their worse case calculations.

During the shortages two weeks ago the wind farms were producing at 8% of rated capacity, and that isn't an unusual amount for this time of year. The wind farms do fantastic in the spring/fall when we don't need much energy, but then in the winter and summer they are frequently less than 15% of rated capacity.

Yep... you need 100% of use covered even if it's a windless night (no solar, no wind), and then turn off (eg hydro, gas) when you have both wind and sun.

Puting just wind and sun infrastructure means constant blackouts.

Or pumped storage. People keep telling me that there arent enough geographical locations for it but all studies Ive seen refute this. E.g. https://www.anu.edu.au/news/all-news/anu-finds-530000-potent...

Never mind that nuclear isnt dispatchable either (ramping down production can be done but it basically just wastes energy) and also needs to be paired with storage or gas to work.

Depending on who you talk to and your definitions, nuclear is generally considered dispatchable. The only definitively non-dispatchable sources are e.g. wind and solar where they cannot guarantee power levels and availability during the dispatching window. It is a bit of a spectrum - some sources (i.e. nat gas plants) are considered highly dispatchable because they can be spun up and down and adjust power levels very quickly (i.e. load following).

There is not enough geographical locations. Sorry to bust your dream, but not every country in Europe is Switzerland or Austria with useful valleys and rivers in them

At this point the same goes for nuclear, if a reactor is offline you need some coal and or gas backup. Unless you have hydro.

If you only have one reactor, sure... if you only have one solar cell, even a bird can stop electricity production. But, if you have multiple/many nuclear powerplants (such as france does), you can turn off one reactore and the others will take the load, fix the first, turn it back on, turn the second one off, do whatever, bring it back on,.... With solar, when it's night, it's night for all the solar cells in your country, and when there is very little or no wind, this can be true for whole regios or even your whole country again.

You don't have more reactors than you need. France gets 9% of its electricity right now from gas.

First, as others have pointed out, Nukes can be throttled, they don't tend to be because its not economically efficient (basically the entire cost is front loaded in the construction).

But there is a _HUGE_ difference between simply not having the capacity because the wind isn't blowing, and having to predict 12-24 hours in advance what the energy load is going to be. In the latter case its entirely possible to spin a nuke or coal plant up in anticipation of the need.

Plus, if you overbuilt nukes, and could guarantee a consistent energy overproduction then you could build all kinds of energy soaking plants, to say desalinate sea water, produce various liquid fuels, etc as well as drive the acceptance of electric cars with free night/weekend charging schedules.

This is actually one of my issues with the Biden administrations "we need more charging" stations drive recently. Charging stations are a problem that will fix themselves when people can look at electric cars and actually say "i'm not driving a coal burner around" and its a no brainier to charge this thing vs buying gas. Until recently that calculation isn't clear, your playing around the edges with the numbers, and the best cases aren't even 2x. Continent wide off peak free and clean electricity for charging makes it obvious which is better. When that happens gas stations will scramble to install charging.

> But there is a _HUGE_ difference between simply not having the capacity because the wind isn't blowing, and having to predict 12-24 hours in advance what the energy load is going to be. In the latter case its entirely possible to spin a nuke or coal plant up in anticipation of the need.

It seems like you think that no wind or wind comes as a complete surprise and that they don't use models to estimate the wind for the next day? How do you think the day-ahead prices are calculated? 1GW missing is 1GW missing, it doesn't matter if that's wind or nuclear. That mismatch needs to be accounted for.

The point isn't that they aren't modeling it, the point is when there is a shortfall, they can't just tell the wind to produce more, they have to find more traditional sources, and when those run out... well... calls for conservation and rolling blackouts when those fail.

And they can and do, tell traditional plants they can't go down for maintenance, or that they have to come back online.

I don't get how this relates to my original point. Do you agree or disagree that both wind and nuclear need fossil fuel backups if there is no hydro that can fill in for the missing capacity?

No, because I think nukes can be throttled and designed for load following, particularly on the timeframes required for grid supply.

Are you suggesting to run the current ones at lower than optimal output and only run at optimal output when necessary? That doesn't sound like a good use of capital.

Vs having 100% of your worse case capacity in gas plants sitting around idle?

The solar/wind numbers aren't even remotely comparable at this point because no one factors those in until the power bill shows up, then somehow the price goes up as we add more "cheap" renewables.

Then each kWh costs even more. But they already cost too much.

Ok sure, ignore the CO2 costs... Because without nukes your going to get carbon production, and with energy usage growing year over year, CO2 usage will continue to increase. If your lucky it might go down for a couple years before resuming the upward trend, but that doesn't even look probable at this point since countries are killing their carbon free sources (nukes) faster than they are growing the percentage of CO2 free energy production.

And those nukes are going to look downright inexpensive if/when we ever get the point of consuming more than say 2/3'rds of the yearly power from wind/solar.

So, keep wishing for that magic renewable bullet and keep watching the gobal CO2 increase and getting pushback from people who aren't happy about rolling blackouts, or their energy prices going up and up and up.

If the greenheads got out of the way and stopped trying to kill every nuke plant the engineering and construction prices would fall significantly. Its not like they are actually that hard to build, we did it in the past, and places that aren't Europe and the US somehow manage to do it for significantly less too. AKA, the price is a political problem not a technical one.

Each nuke project greenlighted means burning coal for ten or more years while building it. The money spent just on that coal over those ten years would have paid for enough solar to displace it all, immediately, and then more after.

The money handed over to the contractor for the nuke, meanwhile, could have instead built out several more matching solar and wind farms, that would also come online in just a year or two, displacing that many times more CO2 emission.

So, each nuke plant started, diverting money from building out renewables, brings climate catastrophe nearer. Coal interests are lobbying now for nuke projects, because it means a continuing market for coal while they are built, vs. wind and solar that displace coal market share immediately.

? This sounds like fiction, have any actual numbers?

Even nuke prices outside of the US and Europe are barely more expensive these days than the cost of wind and a gas plant, making them actually less expensive when the gas plant fuel costs are factored in over its lifespan.

Spend some time with: https://duckduckgo.com/?t=ffsb&q=nuclear+construction+prices...

For starters lets look at: https://www.vox.com/2016/2/29/11132930/nuclear-power-costs-u...

which has such wonderful nuggets as: "A breakthrough came in 1963 with GE's contract to build a low-cost light-water reactor at Oyster Creek, New Jersey. By the late 1960s, overnight construction costs for new reactors had dropped to $600 to $900/kW in today's dollars — cheaper than modern gas plants. Atomic energy was on a roll."

The cost is basically bullcrap, even now, if you look at the prices for American and French companies to build plants in China sure its more than a wind farm, but actually about the same per KW when you factor in the gas plant needed to backup that wind farm, nevermind solar...

https://www.world-nuclear.org/information-library/country-pr...

And those are basically one off, if you think the Chinese are going to be paying those premiums for their own home grown designs I think your probably missing the past 25 years or so of Chinese economic theory, which is summarized as learn how to build it utilizing foreign companies, frequently at great cost, build a few experimental copies/etc and then start stamping out home grown designs for a fraction of the cost. That is after all why your solar panels are so inexpensive. At least where I'm at though, the price of solar hasn't really declined much over the past 15 years or so, because the labor costs now account for probably 2/3th+ of the installed system costs. Going from a $1.50 a W panels to $.60 panels while the install cost climbs at a faster rate hasn't been a winning combination.

Let us know when you have installed a nuke in your house. Comparing cost of rooftop solar here to utility-scale nukes in countries with lax safety regulations and enforcement says more about you than about utility costs.

We already have gas plants, and would anyway need them regardless, so including them in the cost of the renewables is special pleading.

These tricks do not fool anybody. Shame on you.

There's a story making the rounds here about a family who is out $11k because they bought a used ford fusion. Within 6 months of buying it, the battery died. If they could buy a replacement battery, it would run them another $14k, but they can't even do that, because the batteries for the fusion aren't being made anymore. The dealer offered them $500 to take the car off their hands.

I think the current median price for a new electric car is something like $67k. Buying used is a bit like taking the gamble of not having health insurance- maybe you'll be fine, maybe you'll get a life-wrecking bill.

That is ugly, at least in the US the car manufactures are under no obligation to provide parts beyond the warantee period, but they generally do, because its very profitable. That said, I tend to drive the wheels off my cars, but since my first, I've tended to buy very popular models, meaning that there is a nearly unlimited supply of junkyard parts. I would expect that is a partial solution to their problem. A wrecked car probably has a fine battery if it wasn't damaged in the wreck. That is the usual answer even with ICE cars because sh*t happens to them too, blown engine/transmission/whatever can easily exceed the purchase price/value of the car.

OTOH, there are probably various borderline amoral ways to deal with the problem by shifting it somewhere else.

Except that engines and transmissions almost never cost as much to repair or even replace as batteries (which are getting more and more expensive over time as supply chain and materials shortages increase). They are also repairable by more people, including the actual vehicle owner.

That is largely though because people aren't buying new engines/transmissions from the manufacture when they fail. Instead there is a robust remanufacturing industry which takes the old broken ones, and basically repairs them and sells them back through the auto parts supply chains.

So, while its true a new battery pack costs a small fortune, most of them probably aren't completely useless when they fail. If car packs are anything like power tool packs, old laptop packs, and the like then they generally go "bad" when a cell or two get far enough out of spec that its not safe to charge/discharge them as a pack because the charge controllers aren't fine grained enough to manage each cell individually beyond monitoring. The internet are full of DYI projects where people collected packs with 18650 batteries, tore them out, quantified them, and then rebuilt them into packs with matching cells.

I'm betting this holds true even for car packs not using a battery standard like the 18650s. Given a pile of back packs, they could be torn apart, the cells rematched, and then rebuilt into remanufactured packs. Sure they won't be 120% or whatever over their original capacity like the new ones, but a battery pack with 85% of its capacity sold for 1/8th of the price of a new one will likely have a market.

BTW: Couple years ago when I was in yellowstone, Toyota had a big plaque on one of the visitor centers about how they were reusing prius batteries for solar backup.

https://www.greencarreports.com/news/1098326_used-toyota-hyb...

So, even when they are at 50%/etc of their capacity they have a use in stationary applications.

I think it is a pretty bad sign when the cost of replacing a part is higher than the cost of the whole thing.

I suspect this will continue to be true for any used EV whose MSRP (i.e. brand new price) is under $30k. If true, then EVs are the vehicle equivalent of single-use plastics.

btw... there is _HUGE_ difference between it is possible to do something in theory than in praxis, there is NO nuclear reactor that could scale in short periods of time. france which is _HUGE_ on nuclear still has at least ~7-12% of gas (depending if they can use solar and wind or not) and they still import power. so it's basically bullshit that it's just economically efficient, it's simple not possible, else it would've been done, but there are only load following plants and not more and there are still no small plants or fusion plants or some other stuff.

Its not theory, the US navy doesn't run gas backup plants on the subs/carriers. Their reactors follow the load, and its not an insignificant variation, since propulsion is on demand and in the case of a sub, likely the largest consumer.

1. they use diesel generators as backup (they need to and it's a safety net)

2. they follow the load but they also pump out unnecessary energy (heat) into the ocean (you can't do that with normal reactors, BECAUSE you don't have infinite cooling water)

3. it's way smaller and the load is way smaller < 100 MW which of course makes load following way way more easier

Emergency use != load following.

The fact that sometimes energy needs to be dumped != system cannot be throttled. And a huge number of nuke plants are already on the coast so, they have just as much water as a sub (not that this really matters).

And I suspect size isn't really a factor here, its more about time variance, which I suspect is higher on a ship than its is on the grid. The grid wouldn't be swinging its power usage as drastically or completely as one would expect from propulsion onboard a ship.

This is more about whether its possible to run the reactor at say 50% rated capacity for long periods of time, which AFAIK naval reactors can do, and to a certain extent so can some civilian plants, rod placement controls not just criticality but in the end heat generation, which translates to power output.

The best time to build a nuclear reactor is 20 years ago.

The second best is today.

Perhaps the best time to build one is 20 years from now with the hope that by then someone else has sorted out how to do it without exceeding an already enormous budget by a factor of four as with Olkiluoto (so far, it's still in trials that it failed on the first attempt). Plus, to get out of gas you'd need tons of storage anyway as current nuclear reactors aren't very reactive, that's what gas plants do. But if you do invest in storage and make that work well, might just overbuild renewables and limit nuclear reactors to specific niches.

> how to do it without exceeding an already enormous budget by a factor of four as with Olkiluoto

How about imprisoning the responsible for the exceeded budget?

How does Microsoft/Walmart deal with someone promising a thing will cost X but then costs 4X?

They shut the project down before the sunk cost fallacy grow too large.

With nukes, the cost and schedule overruns are the whole point. It would not be approved at all without prospect of those, as payoffs for gatekeepers.

That is why SMR will not find traction.

I find it amusing how the phrase "overbuild renewables" is thrown about several times in this thread as if it's a good thing. It's not. You must acknowledge that the materials and supply chain required for the wind/solar hypothetical simply don't exist.

Reply on news.ycombinator.com