Comment by jillesvangurp | Hacker News Reader

Comment by jillesvangurp | Hacker News Reader

Parent Full thread

One thing that's often overlooked in statistics like this is roof top solar because it's just not that easy to account for. In places like Australia, where double digit percentages of homes have solar and where building codes are actually being changed to require solar panels, this is a non trivial amount that is putting a lot of pressure on energy suppliers to adapt. Effectively whole states are running on solar when the sun comes out (which it does a lot over there). Whether they like it or not, demand for grid electricity drops a lot whenever these panels are producing. And of course a lot of people are installing batteries as well. That must be happening in the US as well and it must be having some impact.

https://www.seia.org/solar-industry-research-data This article seems to suggest that the amount of installed solar has doubled in the last four years and that the pace is accelerating. Also it states that the solar market expanded by 40% last year.

Keep in mind that many US states are actively implementing legislation to disincentivize solar installation.

Not disagreeing with your statement, just pointing out that there's many headwinds to a greener energy future.

Sure, but is that slowing down installations or just setting us up for a correction some time in the future when people install solar anyway? It seems to me that the market for rooftop solar is in any case supply constrained, not demand constrained.

And I can't imagine this being very popular with homeowners or the politicians that represent them, regardless of their political color. So, there might be some backlash against this as well at some point.

Correct. Legislation to disincentive rooftop solar works until battery economics improve sufficiently, at which point the install velocity accelerates because you can sidestep unfavorable net metering or time of day pricing. You’ll then consume your own solar during production, consume it from batteries when not producing, and relying on the grid for shortfall. The next fight will be the legal ability to disconnect from the grid and be self sufficient when they attempt to make connection charges onerous and mandatory.

Tangentially, at scale, this is referred to as a utility death spiral.

https://eepower.com/news/71-of-u-s-utilities-see-the-utility...

The future of some electrity giants is "peaking" (cloudy / windless days and nights) and electricty transfer. They need to adjust for the future, instead of cry like big babies.

you could imagine governments charging a yearly tax for having solar panels installed, analogous to the property tax many charge on real estate

Spain tried something similar and reversed course eventually.

https://www.reuters.com/article/us-spain-politics-electricit...

that's good, i'd only heard about the initial boneheaded policy

I have solar on my roof, but I had to negotiate a net metering agreement with my utility company. I have friends saying the same utility company is now pushing back on new agreements. It seems that stupid issues will need to be resolved before solar hits a critical mass.

That said, batteries help.

Net metering is a bad deal for utilities, though. The “delivery” part of the cost of electricity is more than half; it accounts for the maintenance of the grid etc. People dumping energy into the grid doesn’t lower that cost.

This is an important point that applies more broadly: anything that lowers electricity costs for consumers is a "bad deal" for utilities.

Since in most areas, utilities are monopolies that are not subject to competition, they should not be allowed to deny cost-saving tech advances just because it lowers overall system cost and their overall percentage of revenue and profit goes down.

The thing is, there are issues with handling a grid that go beyond the money aspect.

Balancing a grid that relies on a few production plants is already a non-trivial endeavour. Changing the network is usually a slow process. Adding a new power line to fix some topologic issues can take years to plan and millions to build. Having a robust grid means grid operators have to know details of the network, and be able to know how to reroute power when a line becomes too close to its limits or (worse) when a circuit-breaker opens. Studying the current network takes time.

That's why adapting to frequent changes isn't realistic, purely from the carrier's perspective, without even looking into power production aspects.

Thank you for saying this out loud. Let me preface by saying that I am not a shill for big electricity companies. The cost of electricity, including green energy, and "net metering" needs to including the cost of transfer -- basically "big wires". Most people are delusional about the cost of transfer. One idea: Force generation and electricity transfer to be separate companies. Today, they are frequently a single company. This might help to create more realistic "net metering" pricing. Also, it might help to have some gov't rules for "net metering" pricing contracts.

Edit

So typical here: Your comment is being downvoted. It still stands an important point to consider.

by 'transfer' do you mean distribution, transmission, or the kind of grid operation carried out by isos and rtos

with the right pricing incentives, rooftop solar ought to decrease the amount of transmission capacity required rather than increasing it, by generating the required power closer to the point of use

Who transmits electricity to the house at night? The lines will be sized for full loads.

transmission lines do not connect to houses, so it is not clear that you understand the distinction i am asking about

There are various lines at various voltages. Some are 18k+. In commercial manufacturing may be 13.8k, which is further stepped down to 480 and 220 and 110.

The point is that unless a residence can independently provide for their own electricity every hour of the day, then the generation and distribution will be sized for full load.

it still seems that you are unaware of the distinction i was focusing on between transmission systems and distribution systems

a residence that can provide some of its own power during peak hours does not need distribution, transmission, or (utility-scale) generation sized for full load; a neighborhood that can provide some of its own power during peak hours does not need transmission or (utility-scale) generation sized for full load, just distribution. this is a much less onerous requirement than the one you erroneously believe to be necessary

Who can generate at peak load, reliably, every hour of the day? Some areas have peak load in evening, when PV output is reduced.

Most of the components don't have a meaningful "maximum load". There's a load with 40-year life, a load with 10-year life, and a load with 160-year life (and obviously this is a continuous curve). It's okay to spend some time at 10-year life load if you can spend more time at the 160-year load than the typical model expects.

things like transformers have a 1-minute load which, if you don't exceed it, they'll last 40 years, but if you exceed it significantly, they'll often catch on fire and explode within a few minutes

it's true that there's probably a load right at the rated load which will gradually cook the transformer over the course of 10 years or so, but that's a pretty narrow window

that's the kind of capacity that needs to be expanded

It's a far wider window than you think.

i can't make any kind of sense out of your comment

Doesn't really change the point. Current infrastructures are rarely built to have distribution networks feed back into the HT grid. They are also not as monitored as the HT grid is.

This may change in the future, but until some politician cares enough and passes the cost on to the taxpayer, there's going to be friction for these changes.

> The “delivery” part of the cost of electricity is more than half; it accounts for the maintenance of the grid etc. People dumping energy into the grid doesn’t lower that cost.

This sounds like a pretty good reason for utilities to be a public affair.

The transmission corporation in the state where I live India has had net metering option for a while now. They recently introduced grid maintenance cost chargeable to those exporting solar power to the grid. Despite that it's no brainer for us here to install roof top solar with net metering.

I live in the southeast and was quoted 35k for a 7kw Solar and 36k for two power walls. With tax credits,~50k

I’m in BC, Canada and just pulled the trigger on 14.56 kw if ground mount for $42k CAD, tax included. Wondering why Solar is so much more expensive (ignoring the batteries) in your area? I’d think it would be a lot cheaper with more competition.

Going with Longi bifacial panels - does the US restrict or add huge duties on Chinese panels?

Tariffs are a big part, but labor is a significant factor as well.

GP did not mention the mounting option, so I assume they were going for a roof installation. Which has to be a significant markup vs ground mount. I am only partially handy, but even I could throw some poles in the ground and a frame to hold solar panels. Doing the same for a roof would be significantly beyond my abilities.

From the quotes we received ground mount is usually quite a bit more expensive per installed watt. The frames cost more, there’s longer runs of cabling, etc.

Huh, never would have guessed. I assumed cabling was basically free, and the only difference would be the solar mounts were attached to poles vs a roof.

There was a recent experiment[0] where a company is going to start putting panels directly on the ground without any mounting hardware.

[0] https://news.ycombinator.com/item?id=33926683

Just as a comparison, I did my own 6.7kWh ground mount array, so I can tell you that the materials costs (to a homeowner) for that were about US$13k. That was in 2020 - not sure how those costs have moved since then, but a significant part of the cost you're being quoted is labor.

i'd be interested in a rough breakdown of the costs if you happen to remember

solarserver's photovoltaik preisindex makes it seem like that's about 10 percent pv modules and the rest is either retail markup (or tariffs) or balance of system; is that true?

Alas, I only remember that I paid what was then wholesale solar (OR) almost exactly $10k for the panels, racking, inverter, and the little optimizers that sit under every panel, then about $1k for the iron tubing for the rack, about $600 in concrete, and another $1500 on sundry electrical supplies (conduit, wiring, junction boxes etc.)

thank you!

Have you calculated the savings each month in your electric bill and compared that to the cost of a loan for the solar? Maybe this is a bad time because of interest rates but my sister has a two story house in NY with solar and is saving between $100 and $150 a month.

I don't really understand the advantage of rooftop solar over just putting the same number of solar panels in a field somewhere. Id think it would be a lot cheaper to install the panels in one-place then having to distribute them to a bunch of differently shaped and oriented roof tops, and cheaper to have a few large inverters instead of a bunch of small ones.

I suspect it's political.

Technically, there's also distribution losses (just because we can do HVDC, 5% losses per 1000km, doesn't mean we want to spend the money for it); but I think it's mainly to avoid politicians decrying prime farmland being used for a thing they don't think looks as pretty as sheep grazing on grassy rolling hills.

Depends on the country - some fields are better used for construction, farming or just being left as a natural habitat. And the roofs can't be used for any of those three things. So...

But it depends on the country. Some countries have a whole lot of "low use" fields, like natural or industrial deserts. Others don't.

1) you would use up less nature as no new land is repurposed 2) you produce it in close proximity to where its used

Rooftop solar can help cool roofs/attics by a small amount.

Electrical transmission and distribution costs dwarf the cost of solar generation, 1x-5x the cost of solar generation, and you have to beef all that up for utility scale solar.

Also, the interconnection queues for attaching your project to the grid are getting astronomically long now, and this is one of the biggest impediments to new renewable generation. One of the most valuable assets of a coal generation sites is its connection to the grid; all the rest of facility might have negative value, but the value of the connection can offset all that.

Utility scale solar is great, until you consider how to connect it up to load.

I'm honestly surprised about transmission and distribution costs dwarfing other costs. In my country, it's about 30% of power cost, and power generation costs here are pretty cheap, considering they've been amortized for a while now.

Are you working in an area with a low consumer density?

> Also, the interconnection queues for attaching your project to the grid are getting astronomically long now, and this is one of the biggest impediments to new renewable generation.

Yeah, that's unsurprising. Power Grids are not something that adapts fast to topological change.

Those costs are paid either way though, aren't they? At least, it seems that close to 100% of houses with roof-top solar are still grid connected, so I don't really see how they'd achieve any savings related to transmission and distribution.

With no behind-the-meter solar, the peak for T&D is during midday.

With behind-the-meter solar, this peak is capped, and if the local solar is allowed to power the rest of the distribution nod with net metering, then that means less transmission too.

T&D must be sized for the peak, so lots of residential solar means a lower peak and lower T&D costs.

Of course, this goes agains the utility narrative, because they can't admit that or else their regulators might have to allow such residential net metering. Instead, utilities try to frame the argument that solar customers are taking away from their neighbors or somehow not paying their fair share.

But when evaluated as a whole system, using massive distributed solar along with utility scale is far cheaper than building it all utility scale due to the T&D savings.

A decent reason to have rooftop solar is that it reduces daytime distribution node demand which reduces wholesale commodity prices and takes pressure of the electricity delivery infrastructure (wires, transformers, etc). These are real costs we all pay in terms of capacity and distribution.

>I don't really understand the advantage of rooftop solar

I see many advantages, among them:

- You don't have to find new land; roofs are ideal for solar.

- No need for a huge project, financing, bureaucracy. Done in weeks.

- Costs are low and spread over tons of landowners instead of requiring fundraising and negotiations

- You can have literally thousands of workers busy creating solar power at the same time.

- You get geographic spread and will be less affected by clouds over your central solar plant.

- No single point of failure

- Less need for large and expensive long-distance powerlines

Rooftop solar puts the power to solve a problem in the hands of the people experiencing the problem and eliminates the need for a megaproject. This has turned out to be really powerful, rooftop solar has grown extremely quickly.

An interesting selection of replies you got. Here's one that I'm sure Americans can appreciate:

If the panels are in a field somewhere, an energy company gets the money.

If the panels are on my roof, I get the money.

The American would of course insist that the energy company has a right to reach into their wallet.

You don't have greedy third parties exploiting monopolies over the electricity you consume.

By all means, it should be much cheaper to make large solar farm on some suitable place and buy the electricity from there. The fact that it isn't can't be explained by technical factors.

rooftop solar has the advantage that it doesn't depend on installing new transmission lines to get a grid connection

this is especially important in backwards countries like the usa with their so-called license raj preventing modernization

since most household energy is used for low-grade heating and cooling (refrigerator, air conditioner, oven, clothes dryer) i think thermal energy storage is likely to be a crucial enabling factor; mit's solar house used phase-change materials, but i suspect thermochemical energy storage is a better option

(why thermal energy storage instead of batteries? in the limit it's about three orders of magnitude higher capacity for a given price)

> ...have solar and where building codes are actually being changed to require solar panels...

Do you have a source for that? It sounds kinda silly. The maintenance + OH&S aspects of solar panels are nontrivial and it doesn't make sense to mandate them on residential homes.

I've always assumed that once the economics make sense it'll be easier to build a massive solar farm and let people use the grid as usual. Much less risk of people falling off roofs,heavy objects falling off roofs, wiring being misconnected, weird maintenance problems, managing the ebb and flow of energy, etc. I don't want to have to look after my own panels.

I will install solar(not in US) but AFAIK batteries are not yet cost effective, Am I wrong and someone invented some cheap batteries? I expect that a company could be more efficient building giant batteries(chemical, gravity, etc) then a household buying something and maintaining it.

Batteries are not cheap, but they are cost effective based on local conditions.

Time of use billing. The battery is generating ROI if it can arbitrage energy over the day.

Net metering + solar. If net metering is not available, or sufficiently discounted, and your energy usage is not high enough during the day to use all your solar power, then batteries generate ROI.

Grid stability. A battery effectively is insurance against outages of a few hours. With sufficient solar and rationing, that may extend to week long outages (think natural disasters).

Yes indeed, even a small battery that allows you to shift the late afternoon usage of your house away from the grid will generate ROI by saving you electricity cost compared to a lower feed in tariff during the day.

I think a solar system that also gives you independence from the grid is much more expensive, the program that we are using to get some subsidies do not include this extra features , so if power is out even if we have solar we still be out of power. Maybe in 10 years or realistically more a more advanced system would be worth it for my situation.

Reply on news.ycombinator.com