Hacker News Reader

Comment by vkou | Hacker News Reader

> Why bother with transporting energy when you can just generate it where you need it?

For the same reason every household no longer grows, harvests and threshes its own wheat, bakes its own bread, and why Mao Zedong's Great Leap Foward idea of building a blast furnace in every village, in order to increase the country's steel output was an utter failure.

Power transmission is cheap, and economics greatly favor utility-scale deployments. You also get significantly less need for wasted peak capacity when multiple power producers can pool together into a grid.

View on HN

> For the same reason every household no longer grows, harvests and threshes its own wheat

This feels like a weird point to make when solar power is as popular and growing as it is.

Localized power generation is not only here now, we already have programs to tie your localized power generation into the existing power grid and you get paid for it. I don't see how this system couldn't work the same way.

It's popular and growing because of credits and subsidies. Remove all credits and subsidies, and I can build utility-scale solar for less than half the cost per KWH than you can install panels on your roof.

There's way too much human labour involved in getting someone to drive to your house, climb onto your roof and bolt panels to it. In that time, that same worker could set up a dozen similarly-sized panels when building out a utility solar farm.

And yes, if your time is worthless, and you don't value your neck, you could DIY, and save on some of those costs.

... But you'd still be tied to the grid (and paying grid fees), unless you are ready to invest $XY,000 for a massive battery bank... That might still leave you without electricity during a period of low generation/high consumption.

> There's way too much human labour involved in getting someone to drive to your house, climb onto your roof and bolt panels to it. In that time, that same worker could set up a dozen similarly-sized panels when building out a utility solar farm.

Where do you put the solar farm?

Unproductive farm fields.

Literally anywhere that's not prime real estate. There's three orders of magnitude more of that kind of surface area on this planet than there are single-family home roofs. [1] I think we can figure something out.

[1] 2 billion 'single-family' homes, 800 square feet of roof/average [2], ~50,000 square miles of roof space. Total land area of the Earth is 57 million square miles. You can take your pick of which 50,000 of it can be used for utility solar...

[2] This is a large over-estimate, reality is much smaller than that.

If you take Europe as an example, every area that is not "prime real estate" is either agriculture or protected forests. Transport of electricity over very long distances is not easy, see https://www.pv-magazine.com/2021/05/18/geopolitical-impact-o... for a few of the challenges.

The interesting question is what happens if battery prices continue to decline so that $XY,000 becomes $X000 and thereby less than grid fees.

Low generation can be solved with an inexpensive gas generator for use in emergencies the likes of which might see three days use in a year.

If battery fees get lower than peaker plant-related grid fees, then grids will shut down peaker plants and instead buy warehouses full of batteries.

Aren't some significant fraction of grid fees the grid itself? The land, paying linemen to repair storm damage, the administrative apparatus to negotiate with everyone and collect payments.

You still need all of that if you put all the batteries together in a warehouse, not if you put them in each person's basement.

Those costs are still significant lower than building XY square feet of basement, that you now can't use for anything, because its space is physically occupied by electric infrastructure.

Again, economies of scale.

Don't you still need the same amount of space at the warehouse?

Tesla Powerwall 2 appears to be <5 cubic feet. That doesn't seem like a lot.

I don't think the grid can ever go away. Regional solar is just too variable. You might get backup for a day, but storage costs scale linearly with duration.

>And yes, if your time is worthless, and you don't value your neck, you could DIY, and save on some of those costs.

This is such a cop out. How I spend my time is up to me. If it's a DIY weekend project being worked on a weekend that I had no other plans, then it's not really a cost to me. Sure, professionally, I have my hourly rate that determines my "worth". However, I do not get to bill those hours 24/7/365. Even playing along with your premise, if I'm playing weekend electrician, I'm not a master electrician making the same rates as my other job so a 1:1 correlation is just a lame argument.

How you spend your time is fine. Most people lack the expertise to maintain a solar panel. This becomes an issue when you die or sell your house and the new owners don't know anything about maintaining solar roofing. Over time, this will become a massive cost center especially since solar panel installation have a painful lack of standardization, especially DIY panels, which drives up costs immensely.

>This becomes an issue when you die or sell your house and the new owners don't know anything about maintaining solar roofing

I'm setting aside the "when you die" part, but acknowleding its presence.

However, the new owners get a small tear and a tune from the world's tiniest violin. They know the panels are there when they buy the house. It's not a surprise at closing. If they decide to buy the house with that knowledge, then that's on them, their realtor, their inspector, etc. It's like saying someone that buys a full size truck and complains they didn't know the expense of fuel.

At some point I'm going to pay someone to come out re-shingle my roof anyway.

Sure, but even shitty tar paper roofs last 20 years. Stone and metal roofs can last much longer.

Solar power is inherently distributed by the sun shining everywhere on Earth.

Other forms of energy, perhaps with the exception of wind power, do not work like that.

> This feels like a weird point to make when solar power is as popular and growing as it is.

Funny (ironic?) thing about most the significant residential solar installations in my hood; they exist because they're grid-tied and generate revenue for the households.

If there were no grid for these homeowners to sell their excess power to, they wouldn't have bothered with installing and maintaining solar panels. It's not like there's zero risk involved, hell just having thousands of dollars of solar panels sitting on your roof alone is a big fat sign announcing "well-heeled folks reside here, break-in and pillage when absent". Not exactly the greatest thing to broadcast in the predominately low-income desert regions solar is most applicable to.

Utility scale solar has a MASSIVE cost advantage.

Also, the vast majority of houses can’t sustain their energy use off of rooftop solar alone.

Ive read that creating enough food for a family of 4 from a home garden requires something like 8 square meters of land, chickens, and loads of labour. It’s non trivial, and if you make a mistake you go hungry.

Assuming some putative ideal future Mr Fusion, plugging it into the wall would be a completely different proposition, require relatively little space, and zero household labour.

Considering the massive infrastructure and street furniture required to distribute electrons, the unit economics of home fusion would need to be terrible in order for centralisation to remain competitive against the significant benefits for reliability and decentralisation.

> 8 square meters of land

Pretty sure this is off by a few orders of magnitude.

Why? Staple crops can be planted very close together and there are many crops that can coexist on the same plot.

Crop nutrients are obviously a concern, but if you're only trying to survive for a couple of cycles seems totally feasible and could be extended artificially.

Have you actually tested this in any way? How many calories/year do you grow in each square meter of your own garden?

In the real world, people doing this for a living find that they need literally 100 to 1000 times that much land to feed a family of four.

Have you actually tried to grow stuff?

I grow a lot of stuff and have for years and it’s an excessive amount of labor, you only get certain food during certain times of the year, sometimes for only for a few weeks, and that’s assuming you don’t have losses due to pests.

Agriculture is best left to the professionals for feeding societies.

On the contrary. I think everyone should try growing some of their own food. It's not only great for your physical health, but your mental health too! Also having some long-term food storage and water on hand is a great way to weather unexpected circumstances. It doesn't have to be an all or nothing thing. It's sort of like having a generator or battery storage for backup power... If everyone had it, the whole grid (or food availability) would be a lot more resilient. Imagine if all of Ukraine's residents had plenty of food and water in their homes for the next 2 months. That would be huge.

Being prepared by storing two months of food and water has nothing to do with growing your own food.

If your municipal water supply is compromise, the last thing you should be doing with water is pouring it on the ground.

Conservatively each person needs 1500 calories per day and that's cutting it close. That's about 2 million calories per year for 4 people.

Potatoes are one of the most calorie-dense vegetables by weight and by growing space. They have 350 calories per pound, so you need 5700 pounds of potatoes to feed a family of four.

A good potato yield is about 25,000 pounds per acre, so you need nearly a quarter acre of potatoes (1000 square metres) to feed your family for a year.

It's also worth noting that you're not going to be in shape to do much agricultural work on 1500 calories/day.

> Ive read that creating enough food for a family of 4 from a home garden requires something like 8 square meters of land, chickens, and loads of labour.

I'm pretty sure you can't feed a family from what you can grow on a balcony.

/edit: >Research in the 1970s by John Jeavons and the Ecology Action Organisation found that 4000 square feet (about 370 square metres) of growing space was enough land to sustain one person on a vegetarian diet for a year,

https://www.growveg.com.au/guides/growing-enough-food-to-fee...

> creating enough food for a family of 4 from a home garden requires something like 8 square meters of land, chickens, and loads of labour. It’s non trivial, and if you make a mistake you go hungry.

Not even close. Potatoes produce more calories per unit of land than anything else you can grow in a temperate zone garden. Intensively cultivated potatoes may produce 10,000 calories per square meter, but that would be a stupendously successful crop. In other words, to provide all the calories for a family of four, you'd need on the order of 1 to 2 square meters of potatoes per day.

Since you can't really live on potatoes alone, to get adequate nutrition across the spectrum of human needs, you need quite a bit more than that. You could feed a family of four for a year on less than a acre, if none of your crops failed or did poorly. I'd hate to be responsible for trying it on less than that though in the temperate US.

To be clear, I was responding to the OP by comparing the difference between centralised agriculture and centralised energy production.

As everyone has pointed out, you needs lots of land for food. Far less for energy, and so less reason to centralise it if there is a good alternative.

I would be hugely surprised if we have a Mr. Fusion future where you toss garbage into it and get electricity. More likely is that you'd have to buy fuel pellets periodically. It's not free to package them into household-sized packages and ship them to every home.

Oh, god, I can even see the pain... just like inkjet cartridges. "Non-genuine fuel cartridge detected. Please remove and replace with a genuine cartridge."

Tell that to my septic system!

If the cost of generating power drops, then distribution is a more viable model. Especially if you get extremely high fuel density. Also as we've found in California, power delivery can be very expensive.

Reply on news.ycombinator.com