Hacker News Reader

Comment by Grakel | Hacker News Reader

There's no shortage of open area to put solar panels. If we put a (nearly) continuous strip in every highway median, that would be a huge step forward, and still take up practically zero useable space. Even 1% of the federally owned land in the sun belt would generate an enormous amount of power.

View on HN

"if the government were to take over the Sahara Desert, there would be a shortage of sand in five year" -- Milton Friedman (exaggerating, somewhat)

Better to put them on home roofs, and arrange for them to be owned by the homeowners, designed so that the occupants are (literally) empowered when the grid is mismanaged, as in California.

On highway medians control is more centralized, and they would require more maintenance and cleaning, which would be more disruptive in that location.

Rooftop photovoltaic is currently more expensive, less efficient, and surprisingly dangerous to install. Solar thermal designs actually benefit from being on rooftops.

As to dust, economies of scale make a huge difference. It’s rarely worth it, but using a farm truck to spray water across miles of panels is much more efficient than tens of thousands of homeowners doing the same.

It's also unattractive, which should matter to us.

Much of the distinctive beauty of traditional building in sunny areas comes from the roofs.

Solar tiling, which is still not a mature technology, doesn't add danger: we have to put something on a roof, and if installing solar tile takes extra time, it isn't by much. And it looks good.

Less efficient and more expensive? Sure. But as the meme would have it, it's free real estate. And since we're living through a grotesque failure of robustness on a civilizational scale, let me point out that it's more robust as well, particularly combined with on-site batteries.

Why do they count as unattractive? It seems that the "definition" of eyesore is wholey independent of aesthetics and entirely based upon if it is functional and new. Old windmills, steam engine trains, and canals are considered quaint, wind turbines, shipping canals and modern trains are considered "visual pollution". Modern art fixtures often deliberately defy all rules of structure and aesthetics and don't get tarred with this. I only have weird hypothesises as for why.

> unattractive

people used to say wind turbines are unattractive. Culture determines what's attractive, and it can be made to look nice.

I'm dubious about rooftop solar in suburban/urban areas. I think the gain is mostly from tax write offs and arbitrage. Compare hand installing solar panels on a roof with mechanized installation of a large scale solar array.

Just isn't going to compete with stuff like this.

https://www.youtube.com/watch?v=Xk21aAaJL_s

Generation of power at the point of consumption is nothing to sneeze at. It’s also efficient use of space. Taxes can work as incentivizing tool instead of or in addition to generating revenue for the government which is precisely what’s going on here.

What dangers are we talking about? Do solar panels add extra danger compared to just normal roof maintenance work?

To be clear their not that dangerous, but we are still talking extra deaths and accidents. Mostly from increased amount of time people spend on residential roofs. Roofing is about 30 deaths per 100,000 full time workers which is about double the average rate for construction.

Currently instillation is the primary, but not only risk. You also get electrical accidents and in rare cases fires. Plus an increased tendency for people to go onto a roof to check them out and or clean them etc. And some extra roof maintenance due to leaks from improper installation etc.

Personally, I don’t think it’s a major deal at the individual level. However, I thik it should be considered in terms of public policy.

First I’ve heard of solar thermal. Any good links?

Goes back years. Simplest form is a squiggly pipe behind a pane of glass. Water runs slowly through the pipe and heats up.

Think rooftop solar water heaters.

It's most efficient to put the panels in a centralized location with high amounts of sunlight on mounts that track the sun. You get significantly higher yield because of the tracking and the location, and maintenance is cheaper because it's centralized.

The fact that you see far more home installations than farms means farms aren't economically viable, neither is home PV without subsidies, but people buy into the "empowerment" fantasy.

Empowerment isn’t really a fantasy if there are rolling blackouts but your own power is still on. Sadly, not a hypothetical in California.

Im thinking just from a national security/redundancy/risk reduction standpoint it makes sense to move more generation closer to demand. There are lots of single points of failure in our power grid currently.

Edit: this is assuming the solar setup is designed to enable use when power is out; I assume homeowners who care about empowerment would make that investment, but you know what they say about assuming.

> Empowerment isn’t really a fantasy if there are rolling blackouts but your own power is still on. Sadly, not a hypothetical in California.

Seems easier to get a generator or a Power Wall, but either ends up being pretty expensive for handling the 2 hours of rolling blackouts for the first time in 20 years. There are also longer-term outages, maybe several hours every few years? And this is for people in population centers. Living in the mountains is a different story.

> Im thinking just from a national security/redundancy/risk reduction standpoint it makes sense to move more generation closer to demand

I'd rather put the money towards removing those single points of failure.

> Seems easier to get a generator or a Power Wall, but either ends up being pretty expensive for handling the 2 hours of rolling blackouts for the first time in 20 years.

The price of electricity changes over the day along with demand. I don't know by how much, but maybe if the Power Wall could be used in the high-cost hours and charged in the low-cost hours, it would make a difference?

The balance of tracking mounts is not trivial.

They increase efficiency a lot (up to 40% for dual-axis trackers) but they also decrease ground a lot coverage, they increase cost of installation and cost of ownership, and they decrease availability.

It balances out to the positive.

See https://heliogen.com/

That's a heliostat system, not a solar panel system. Both involve the sun but they do so very differently.

It's also a company which wants to sell heliostats, they're not going to tell you it sucks.

Solar farms are in centralized locations. It is entirely possible that no farm has been constructed in that specific area.

> if the government were to take over the Sahara Desert, there would be a shortage of sand in five year" -- Milton Friedman (exaggerating, somewhat)

I know that this is meant to be a humorous comment, however, desert sand is not really usable for most things that we want sand for, for example, it is not suitable for concrete, since it is too fine. (https://petroleumservicecompany.com/blog/could-desert-sand-b...)

This is why, desire being in a desert, Dubai imports its sand! (https://www.bbc.com/worklife/article/20160502-even-desert-ci...)

Solar panels on home roofs aren't necessarily configured to power the house without the grid. Also you would additionally need batteries and related hardware to have a usable source of power at any time, so unless your house is set up for off-grid solar, you still need the grid to be managed well.

Found this out in the house we are renting with solar panels installed by landlord. Bright sunny day, someone hits a pole nearby, power goes out.

There are two kinds of solar: one that adds to the existing signal from the mains, or one that is self-contained. The second kind is more complicated and therefore expensive. I wasn't aware of the distinction before, but I definitely am now.

Aren't there setups that can do both as well? i.e charge battery first and send excess to the grid?

The setups have to be separate (e.g. different outlets for mains and solar) to protect the grid.

Even a small home solar installation puts out enough power to easily kill unsuspecting maintenance personell working on the grid, for example.

So the utility company has to have 100% control over he mains at all time for safety reasons (i.e. your meter must never run backwards). If they cut the power (e.g. for maintenance or repairs), no one must be able to feed power into the grid.

That's why even if you have a battery, said battery can either feed into the grid, or be connected to separate outlets in your house, but never both at the same time.

> So the utility company has to have 100% control over he mains at all time for safety reasons

This is controlled independently by the solar inverters's software, not the utility via any sort of signal. That software is mandated by regulation as you say for the safety of line workers. Recent batteries allow for an "islanding" mode, which allows continued backup power delivery to the house during a grid outage but cuts off the grid connection.

> your meter must never run backwards

... during a grid outage. During normal operation it's fine for it to run backwards. That's what net metering quantifies.

> said battery can either feed into the grid, or be connected to separate outlets in your house, but never both at the same time.

I think this is a technical limitation, since it would require a current divider and 2 inverters, which is a level of complexity not worth it for the application.

> During normal operation it's fine for it to run backwards. That's what net metering quantifies.

Hm. Different countries, different regulations I guess. In my country you have to install a second meter instead.

That's not true, you can have an inverter setup which fails over to a battery backup when the grid goes down. It's also done all the time with natural gas generators.

The setup is more expensive, and it does have to meet certain requirements, but it is definitely an option.

True but, this has been made illegal in many places, California among them.

I don't actually know whether that's legislatively illegal, or simply banned by PG&E; but the effect is much the same.

Meters run backward intentionally when selling solar back to the utility. You need a switching system that will isolate immediately if the grid fails, and resync with the grid AC before reconnecting, if you want to be able to operate on and off the grid.

That depends on the country then - in my country, you need to install a secondary meter instead.

Yes, but any setup that's capable of feeding power back to the grid requires costly safety interlocks to allow linemen to make repairs at the street without your house unexpectedly energizing the line. Some jurisdictions just outright disallow it.

Sure but they are more expensive. You need a rock solid controller with beefy and redundant electronics to make sure you're never going to be connected to the grid when you shouldn't be.

> Better to put them on home roofs, and arrange for them to be owned by the homeowners, designed so that the occupants are (literally) empowered when the grid is mismanaged, as in California.

It turns out that the same California that you criticize now requires solar on newly built homes, just as you recommend, and much to the chagrin of people who feel like the state government is impinging on individuals' rights to build however they please.

> the occupants are (literally) empowered when the grid is mismanaged, as in California.

The California blackouts started around sunset, when solar dropped off the grid. Building more solar without an appropriate amount of storage just makes the problem worse.

The solar panels on roads idea has been tried. Rotting leaves were a big source of its failure.

https://www.businessinsider.com/first-solar-road-france-fail...

It failed because building an efficient solar panel is already hard enough without thousands of cars running over them. Just build a solar-roof instead.

Doesn’t matter. Much of the US roads aren’t anywhere near trees - ie the southwest. Huge amount of roads and solar power potential there too. Also, cleaning crews are a thing that have existed in the past and should again.

Replace leaves with tumbleweeds. Or whatever natural debris we have in the west.

The leaves were causing damage to the solar panels. I imagine other types of debris would do the same.

But I'd be happy to be wrong. Have you asked why this hasn't already been done yet? Or if it has (besides the example I provided), what was the outcome?

Have you spent much time driving in the empty areas of the west? You can go 100 miles without even seeing a tumbleweed.

I've been living in the west my whole life and have traveled between Colorado to California through Nevada, Utah, Arizona, etc 7 or 8 times, so yes I've seen an empty western road before. Tumbleweeds was tongue in cheek but there's plenty of dirt, trash and other flying debris that will get baked onto a glassy surface out southwest.

Well, even in the Sahara there's sand and sandstorms piling it on solar panels...

There's no place that's 100% immune from all dirt/leaves/etc.

Still shouldn't be much of a problem in practice...

I think the key difference is whether the panel is placed at ground level (as in a solar road) or raised (as in most large-scale solar installations where the panels are raised on mounts that can also move and track the sun). The latter is pretty much self-cleaning when it comes to dry debris because the wind would eventually blow them away onto the ground (where the former will collect them forever).

I agree on both. It (degree fouling of the solar panels) is a fact of things, but shouldn't impact things much pragmatically.

I think that we can expect the same amount of accumulation of stuff as we currently get on these roads. Its not like we're out there street sweaping these rural highways, but they're also not covered in debris.

3/4 of Nevada is owned by the federal government and is dry to the point no one wants to be there anyway. Including planets. From your list, you probably haven’t seen the part of the Southwest which I was referring to earlier. As others have said, too, maintenance is going to be a thing for any form of energy.

In the southwest, I'd imagine dust would be the biggest hindrance.

GP said the median, not the roadway itself.

One thing that I don't get about this and related stories: if the road surface is damaged as badly as they say, then why are there no pictures of the damage (the one picture showing damaged modules is described as not taken from modules instLled on the road)? Also, I googled several other stories and none of them has hard facts on why it failed, essentially speculating about different reasons.

Something about this story is not quite right, at least the way it was reported in English language media.

I can’t read the article, but previous articles I have read on the subject indicated that the road authorities who evaluated the tech never found it suitable to install on a public roadway to begin with. Besides the durability/maintenance issues, it also had lower friction than a normal road surface. I think one here in the US gave it a trial run as a sidewalk and even that use was a failure when the freeze and thaw cycle of the ground broke them.

Yeah, land is not at a shortage. The best land for solar is also worthless for other things, so it tends to be dirt cheap. There's enough unused space in the Southeast US to power the entire world off of solar. Obviously that wouldn't work for other reasons, but underscores nicely how land isn't the issue at all.

They would also get dirty at a much faster rate because of all the dirt and soot generated by the vehicles going by.

Just waiting for it to be economical in the nearer term and all of this stuff will magically start appearing. Everyone goes through the motions of thinking about solar solutions to energy, and then they realize how little power they provide and how many they need en masse and then scrap the idea.

Once the math adds up for a quicker ROI, the plans will stop getting shelved and will be presented as novel by different people.

I think we're already there for regular PV. It's getting to the point where a home system will almost pay for itself in less than 10 years.

I'm thinking 3 months would be ideal

Very few things have a return that quickly. In an era of 25 year mortgages, a 10 year payback is very reasonable.

Let's make these one of the few things that have a return that quickly.

That’s unlikely to ever happen; if it’s that cheap to produce electricity the utilities will have long since switched over and cut rates.

Ideal, I guess. Though I don't see how that could ever be possible.

This approach would benefit community solar and remote microgrid projects, as well as geographically constrained places like the UK where the study was done.

> If we put a (nearly) continuous strip in every highway median

Is there a shortage of land for solar?

I was wondering if they can be put up on poles holding train's electric tractive wires and then fed directly to the same wire. The issue i see is dc to ac conversion and lot of variability management.

Vehicles often crash into things beside a highway.

That will surely be taken into consideration from the start, and they will be designed in easily replaceable sections. If a car hits a section maybe it will be disabled but the rest of the cells will be fine and a crew will be dispatched to repair or replace the damaged cells.

Surely it is cheaper to simply place the panel somewhere else than it is to:

1. Pay for thousands of additional broken panels on a recurring basis

2. Pay people to replace thousands of broken panels

3. Shut down highways for utility work to replace thousands of broken panels

4. Design the panels to meet crash safety design considerations that the structures currently in the median were designed to

5. Add additional electrical safety standards unique to a panel array that designed to withstand car crashes.

Really, this is complete silliness.

Planes sometimes fall on people's roofs. Let's not optimize for edge cases here?

This has to be the most ignorant comment ever posted on HN. In a typical year how many aircraft crash into structures? Zero. There are more than fifteen thousand car crashes in the USA every day.

The entire reason medians exist is because they are a crash safety device. Consideration for crashes is not an edge case, it’s the entire point of a median.

Maintenance in the middle of the road sucks ...

Reply on news.ycombinator.com