Hacker News Reader

Comment by szvsw | Hacker News Reader

> “Gaining 1 or 2 percent more efficiency is huge. These may sound like very tiny increases, but at scale these small improvements create a lot of value in terms of economics, sustainability, and value to society.”

It’s so easy to forget this and the massive scale and its relevance at the massive scale of the systems we need to (and are, to some extent) roll out. It also seems promising when these breakthroughs are happening in R&D groups of industry players trying to dogfood it rather than in labs.

At the same time though, it’s starting to feel to me, to some extent, like we have kind of solved solar? It’s everything else around it that needs to advance, particularly grid infra, batteries and electrifying the general class of difficult-to-electrify problems (steel, concrete, freight). I might be totally off-base and blinkered with that assessment.

Edit: I guess I should try to clarify my feeling after reading some of the responses below: it feels like solar tech is not really the limiting factor in renewable scaling, and that advances in solar efficiencies won’t drastically/meaningfully simplify the other challenges/limiting factors we currently face (grid infra/batteries, electrification of mfg, duck curve, etc). Children point out that space and cost savings from efficiency gains in solar may still be significant at grid scale though! Still, this is very cool progress to read about!

View on HN

No, we have not. Half of the current cost is due to installation and labor, so any weight reductions will have a huge impact on cost. Perovskite solar cells have the opportunity to have a 10 X weight reduction.

How much of that cost is related to the physical handling of the panels themselves, and not the electrical installation, though? Weight reduction isn't going to help there.

It’s propagates through the supply chain, loading, logistics, wear on vehicle transport etc.

We should practically get to the point where someone can buy a roll of material at Home Depot and unravel it on their roof, nail it, and plug it in themselves.

The currents and voltages involved are going to make that a non-starter. Do you really want random people messing with those? The solar inverters are also sized that big for a reason.

At least in countries with strong regulations around working with electricity this is simply not going to be feasible.

I've been in the PV business for some time now and seen a person get killed by it. It's not pretty. Still remembering that smell of burnt flesh... Now, to be fair, that was at a 12MW-installation, not on a roof. But still...

Jay Leno talks about installing solar on his house (which he did himself) and commenting that he was getting shocked a lot, bc as soon as the PVs are in the sun, they're making electricity. He said it made handling the units tricky.

Stupid question I've been curious about.

What happens to a PV panel, receiving sunlight, with no load?

Does it degrade or suffer ill effects in any meaningful way? Or does it just have a potential between its outputs but otherwise isn't impacted?

A small current is going to flow internally, but nothing else happens. It's quite normal to have solar panels running with zero load in regions with lots of PV - reason being, that the carriers need to keep their electricity nets stable and have to carefully balance electricity entering and leaving the net.

At least in Germany, every PV installation of certain size (> 30kW peak) is mandated to be able to be shutdown remotely by the carrier if you supply electricity for the net and aren't just using it for yourself. (You get paid the same during shutdowns, just like it were running. Otherwise it would be quite damaging and likely reduce adoption of PV)

Point being: no, it doesn't hurt the panels and is a regular ocurence.

The generated power will be dissipated through the panel as heat, AFAIK.

Which means that in winter, probably nothing, because it's cold, but on a hot summer day with peak sun, the heat might start damaging the cells. How much exactly you'd have to look at studies.

My guess is the output will permanently degrade by a few % per year if the panel is not connected. Might go down to 80% way quicker than normal (25-yr)

> The generated power will be dissipated through the panel as heat, AFAIK.

Solar panels are not constant-power devices. In an open circuit, they will generate their open circuit voltage at nearly zero current (except minor internal leakage), and thus nearly zero power. In a short circuit, they will generate nearly zero voltage, and thus also nearly zero power. To get maximum power out of a solar panel requires maximum power-point tracking (MPPT), where the load is adjusted such that the product of voltage and current (that is, power) is optimized for the current conditions; while significant power can be delivered to a fixed load, there's no real power being generated without a load.

Where's the power going to then? Either it's heating the panels or it's reflected back, no?

Given ~20% efficiency it's almost negligible amount of heat.

The sunlight heats the panel a bit more than usual but there isn't really generation/dissipation going on.

And since heat radiates away at temperature to the fourth power, the increase shouldn't be particularly much.

The thermal power of the sun will heat the panel, to the extent that it is not reflected. But no electrical power (or any power) is being "generated" by the panel, the panel is just absorbing photons like anything else with low albedo.

That's just nitpicking. There is thermal power generated from the electromagnetic power from the sun and GP is right that a solar panel turned off will be hotter than one turned on.

And if that was the original poster's intent, I apologize for nitpicking. My impression was that the use of "generating power" in the given (open circuit) context suggested a fundamental misunderstanding of the behavior of solar cells in this situation on behalf of the poster, and thought I might clarify; and perhaps even if the original poster understood this already, someone else learned something.

Think of what happens to a normal roof tile in the sun: it absorbs solar energy as heat and also reflects some and radiates some away. A solar panel is the same (though a bit more reflective), but when a load is connected some of the solar energy is converted to current instead of being absorbed as heat. Therefore the panel is a little cooler when a load is applied.

There is no power. Power is IV, current times voltage. The voltage will be the rated voltage of the panel at that sunlight level, but the current is 0 (minus some very, very small (microamp) internal currents).

Alternately, power can be expressed as V^2/R. But in an open circuit R is infinite, so again, zero power.

There is ~1000W/M2 of electromagnetic power from the sun, and if it's not turned into electrical power it will be turned into thermal power, thus heating up the panel.

But as noted by sibling comment, that heated panel will then begin to transfer that heat back into its environment, as a function of its temperature difference with its environment.

So as long as manufacturers engineer their panels to be tolerant of the maximum heat at a site (i.e. full sun, maximum temperature), the panels won't be harmed in any meaningful way. They'll just heat up a bit faster than if they were providing current.

Is ther a difference between "electromagnetic power" and "thermal power" here? If a panel is not connected, there is no conversion and the surface is warmed - just like any other surface exposed to the sun gets warmed.

PV panels are just like charged capacitor or a chemical battery with no loads: just holding unused potential differences with no damage to the unit.

Really? Couldn't you just tape a piece of cardboard over it?

I installed an additional 12 505Wp panels by myself last weekend, the panels came with MC4 connectors installed which is pretty standard I think. Hard to get zapped by solar DC juice that way.

BTW: installing solar panels DIY is apparently super easy, as I found out. I have a flat roof and used micro inverters, to make it easier, but I was done in less than a day (excluding selecting the components and layout)

if it were really a problem moving forward and diy becomes the norm(which i doubt is the case) it's pretty trivial to apply an opaque sticker or cardboard covering to the panel during manufacture.

I've already gotten smaller solar panels in shrinkwrap, I could easily see it becoming standard to package it in a scratch proof blackout layer

Why can't they just be covered with blackout film until they're ready to be activated?

I just left mine half in the packaging, but I'm not Jay Leno.

He got a funny story out of it, maybe he could write them off as a work expense.

Or have interconnecting cables, plugs, and sockets that are designed to prevent you from touching the conductors. Yeah you could probably still shock yourself but you'd bascially have to be trying to do it.

That... makes no sense.

I just installed 7.3Kw on my roof, and another 600W on the roof of my Jeep.

You'd have to physically stick a screwdriver into an MC-4 connector to get zapped, which is as smart as sticking one into an electrical outlet.

I don't even understand how you could get zapped plugging in MC-4 connectors.... like, at all.

Do you have a link to that video? I've never thought of Leno as a particular handy guy so it sounds fascinating.

Also I can't imagine him up on a roof. He's 74 and not a pillar of health and fitness.

I doubt this interview was last week. It might have been 10 years ago.

> The currents and voltages involved are going to make that a non-starter. Do you really want random people messing with those?

As part of our daily lives, a great many of us climb into a steel box powered by explosions and packing a 20 gallon container of flammable liquids (and increasingly several hundred pounds of also flammable batteries containing more electricity than an average family uses in a week) and then pilot that box at 80Mph down a strip of concrete packed with other large high-speed objects containing flammable liquids. Occasionally, we run low on flammable liquids in our high-speed metal box and get to refill the flammable liquid container ourselves at a flammable liquids depot, which contains upwards of 40,000 gallons of the flammable liquid delivered by other larger high-speed metal boxes which also share the same strip of concrete with us.

So: I'd expect some product safety iteration here before we get to the "roll out your own solar panels", but no, I don't consider that a non-starter.

To be fair, it didn’t start that way and there are decades of design, legislation and safety regulations around all this, along with infrastructure for licensing/certifying capabilities, tracking and policing capabilities and mistakes over time, insurance, yada yada.

It’s not like that stuff springs up overnight!

the safety of these steel boxes were bought with blood over a hundred years.

and the end user doesn't just cut as much as they need and nail it down - the things are practically disposable appliances at this point.

But you can see fuel. You can't see currents and it's not trivially visible which things you can touch at all, which things you can touch at the same time, which protection to wear, how to deal with the potentially fatal flashing arcs, ...

PV installations on roofs typically have around 10-20kW peak output.

Let's go with 10kW. That's around 25 panels, each outputting 30V with something like 13A. Small installations are typically single-stringed, so you end up with a voltage of 25*30V=750V with 13A DC. That's pretty likely to kill you within milliseconds if you mess up.

There's a reason that stuff tends to be handled by professionals. It's a ridiculous (and pointless) risk if you aren't well educated about it and have some experience.

So don't single string the DIY version into an uncontrolled danger wire. There's several ways to accomplish that.

You could probably make a system sockets that communicate with each other before exchanging any serious power. You could digitally sign cables, and add shielding to make them detect cuts..

I'm not saying it's a future we should want :)

But isn't that kind of how super chargers work?

Of course, until all our grid hook ups are smart, we'll probably need electricians at some point.

Enphase systems work as you describe. Each panel has its own AC inverter, and they from a mesh network and run sanity checks for shorts, etc. before exporting power.

There are lots of boxes in our house that do 16 amps at 120V. Some do 240V, and some do higher than that, and some of those are next to sinks or in high-humidity environments. The voltages and currents for solar panels (especially if you use 240V microinverters) don't seem like a non-starter to me.

That's AC. The panels are DC. That makes a big difference.

That said, the point of "do it yourself" is that you'd nake it less dangerous for ordinarily folk. So the risk of shocks would come down.

What would concern me more is long-term fire risk. If not installed correctly, with the right spec parts etc, proper grounding etc, there's a significant risk of fire. Not immediately perhaps, but a couple years down the road.

Again DIY kits would need to be designed with this in mind.

> That's AC. The panels are DC. That makes a big difference

What's the qualitative difference between 16 amps at 120V and PH v DC? Either is enough to kill a person if mishandled, and yet Home Depot sells breaker boxes over the counter.

750 volts will not kill you in single-digit milliseconds, but rather hundreds of milliseconds (or not at all), and it's probably worth it to run the extra wire to not be single-stringed

And you know how much mockery Oregon and New Jersey get, for believing gasoline is so heinously dangerous as to require trained dispenser operators? Meanwhile the rest of us just pump it into our own cars like adults.

It's funny to look at electricity from the same perspective.

It has nothing to do with safety. It's pure protectionism. The point is to keep small gas stations competitive by imposing labor costs on larger gas stations.

You need a license, and usually some sort of training, in order to operate one.

Always worth pointing out the liquid is not only flammable, but also toxic.

And using the metal box creates toxic fumes that we inhale, which are deadly to every living thing on it.

Homeowners work with 240v and tens of amps all the time across the country. Hell I wired a hot tub breaker panel and a car charger. Safe enough interlocks and it's a non issue

And houses are burning and people are being electrocuted regularly - being not only a hazard for themselves but also their entire neighbourhood. I certainly wouldn't want to live next to somebody who thinks they can handle their electricity installation by themselves.

I doubt running a circuit for a hot tub is endangering the entire neighborhood

I would rather be zapped with a 240v AC current vs 240v DC current.

AC power crosses the zero line twice per cycle while DC does not. AC has a lower ‘let-go’ threshold, but DC contracts your muscles and makes it harder to let go.

You are correct though, if you de-energize your panelboard and have a deadfront cover over the line side conductors and lugs, working inside a panelboard (or on electrical wiring) is safe.

DC interferes with your heart's rhythm much, much less though, due to being constant. AC's frequency easily causes ventricular fibrillations even at low currents and voltages. AC is considered potentially lethal starting at 50V. For DC it's 120V, because it's significantly easier on your heart.

It’s the amperage that kills you, not the voltage. 5000VAC at a 1.0 nano amps is not going to be something you can feel, not even as something like static electricity.

We're talking about proper sources here where the voltage doesn't disappear as soon as you start mildly conducting. So volts and amps will be proportional in this context.

And the other important part is that 60Hz needs fewer amps than DC to be dangerous. https://www.allaboutcircuits.com/uploads/articles/electricit...

I don’t buy that the volts and amps will always be proportional. In my experience, the volts are usually pretty fixed, depending on the circumstances. Like 120VAC in most homes in the U.S., but variable amps — 15, 20, 30, 50, 100, etc…. Or 240VAC in Europe and certain other places around the world.

And if you want to talk about power lines, then the neighborhood medium voltage lines are going to be roughly the same in most places within the same jurisdictions, and distinct from the true high voltage lines that are used for long distance transmissions.

You are not conductive enough to get anywhere near 10% of the circuit's capacity. Therefore, the supply might as well be an infinite amp supply. You, in any particular situation, act as a particular ohm resistor. The amps that flow through you from mains voltage or big solar arrays will be directly proportional to the volts.

If a 120V 15A supply puts 50mA through you, then a 120V 100A supply will also put 50mA through you.

A supply that's "5000VAC at a 1.0 nano amps" really means that it starts at 5000 volts but super rapidly drops to zero volts as it conducts. A household supply is going to have negligible voltage drop by the time it turns deadly.

Edit: The other way to put it is that 99.9% of supplies don't give you a certain number of volts and amps. They give you a certain number of volts and they have an amp limit. If you're not approaching the amp limit then the only thing that matters is the volts.

I get what you're saying, amps are pulled, not pushed. But you should consider ohms law. Hand to foot, with a dry hand, typically is 400ohms. You can pull a lot of amps through 400 ohms if the voltage is high enough

The focus here is the 30-750 volt range with tens of amps, with a secondary mention of neighborhood power lines. For injury purposes, we can treat all of these as having unlimited amps available. By the time there's enough kilowatts going through your body to notice the curve shifting, we're well past the point of wondering whether or not you die.

120v at 400 ohms is greater than the 40ma required to harm the heart and kill someone.

Yes I agree?

My point is that the number of amps you can get from the circuit is irrelevant, it's "more than enough" and that's all you need to know beyond the voltage and the exact way the human is being exposed.

Ah, I apologize. I misconstrued your position

Ohms law calculator:

400 ohms at 110 volts = 275 ma. More than lethal. 30 watts.

400 ohms at 220 volts is double that - a bit over half an amp. Lethal (obviously). 60 watts.

400 ohms at 1500 volts is 3.75 amps - 5.6kw. Enough to physically cook someone pretty quickly.

I'm in Germany, our local discount retailers sell PV for apartment balconies as DIY systems — "Plug and play", even.

If Lidl can do it, why can't Walmart?

https://www.lidl.de/p/vale-balkonkraftwerk-minipv-800-et8-l-...

They're limited to pretty small sizes by german law - they are so insignificant that they're much less dangerous to handle. I'm not up to date on their ROI, but IIR they usually were a rather bad investment and more of a novel toy than a serious and reliable source of electricity.

Essentially, any notable installation fundamentally deals with much higher currents and voltages and as such is much, much more dangerous. Once a certain size is reached, the carrier also has to be involved and professional installation is mandatory, both due to the law and requirements by insurance companies.

At least here in germany. I've been involved with building all kinds of PV installations in bavaria, from 4kwp up to 20MWp. The balkony generators aren't taken seriously by anybody in the industry right now, at least.

> At least here in germany. I've been involved with building all kinds of PV installations in bavaria, from 4kwp up to 20MWp. The balkony generators aren't taken seriously by anybody in the industry right now, at least.

If you're working at that size, I'd expect you to ignore balcony systems regardless of how cost-effective they were.

My point here is simply and only that it's possible to make a system safe enough that an untrained and unskilled member of the general public can just plug it in and use it, which is exactly what was being called for up-thread with this:

> We should practically get to the point where someone can buy a roll of material at Home Depot and unravel it on their roof, nail it, and plug it in themselves.

Germany basically has that (even if it's not in the form of a roll); there's nothing fundamental preventing the USA from having it too.

This is 1/10th of what my house needs (and I'm not living outrageously AND far closer to equator than Germany)

And?

They're sold for apartments, and as DIY jobs. They're designed to fit on a balcony just about wide enough to stand on, and to be installed without needing an expert.

The point of the example is to show that you don't need an expert. It's not even trying to show a specific unit that suits all people, just that one thing, that you don't need an expert to install it.

The voltages are the same regardless, because that's how domestic electricity works. (If you forced me to guess, I'd expect grid-scale PV farms to go direct to a higher voltage than domestic users, but I'm not an electrical engineer).

And it doesn't scale. Plus without proper meter (or CT "limiter") it will clock exported power as consumed.

Don't get me wrong - I'm pretty committed to DIY a decent system, but it's not trivial and what you posted is just a toy.

> And it doesn't scale

The law in Germany may prevent you hooking up ten, but that's not relevant to the point or the market.

Can Americans even hook up things this size on their rental apartment balconies?

> Plus without proper meter (or CT "limiter")

Difficult term to search for, so I'm unsure what that is exactly. I get links about inverters, and I'm sure you noticed this comes with one so it's probably not that.

And this relates to the impact of module mass on (supposedly) preventing DIY installation (despite my example of a DIY installable system) how exactly?

Friendly Neighborhood Handyman checking in here. Average Home Depot customer? Absolutely not unless there's some plan to quadruple suburban emergency services budgets. On the other hand I get pretty tired of sneaking around local restrictions on electrical work. Residential electrical work isn't exactly complex and I can't devote a couple years to working as someone else's laborer to get a cert. I'm perfectly capable of handling 100% of a residential solar install (including battery backup) and it's aggravating af to have to go find an electrician to bribe to get permits and inspections.

Around here you can buy a permit if you’re a homeowner. You then have to set up the inspections and actually do all the work properly, but there are no restrictions like that. The inspecting agency publishes documentation about what to read and common pitfalls even.

Yes, this post seems to be by someone who wants to be paid to do this by a home owner.

It's very common to need licensing to do something commercially but not when doing it for yourself.

I see how you'd think that but it isn't the case. Where I live only licensed electricians are allowed to do any work more involved than changing a light fixture. As an example I've got a 1600 square foot detached shop on property I recently purchased. There isn't six inches of wiring in the entire building that is up to code. My options are ignore it and risk a fire or electrocution, spend >$15,000 to get the building rewired, or spend $3,000 on materials and risk a life-altering fine and/or jailtime if I get caught fixing it myself.

I'd be satisfied if I could simply sit the licensure exam and maybe have to pay extra to do some kind of practical demonstration. Local requirements for residential licensure include documented multi-year experience as an electrician's helper before you can even apply to take the exam.

I can pull my own permits for carpentry and masonry work, structural stuff, but that's about it, and only for my personal residence. Homeowners here are barred from systems work of any kind, and it takes a contractor's license to pull permits when working on someone else's home.

> including battery backup

Ironically batteries is what makes it feasible - I can dump excess into battery instead of paying 3x more for install so I get hooked up to grid in a certified way.

I've had to deal with electrical installations from people who "knew what they were doing so didn't need a license or permit."

And "sneaking around local restrictions" can create quite the fun surprise for unsuspecting workers who need to open up walls, dig trenches, etc.

I hope you decide to play by the rules.

My work meets or exceeds code requirements, every project, every time. If I encounter anything where I don't already have relevant code committed to memory I stop what I'm doing and go look it up. How many tradesmen do you know that can say the same with a straight face? Anyway I'm fine with pulling permits and having my work inspected, I prefer it even when possible.

for an offgrid installation it's feasable to possible to have an idiot proof system. It rise the cost, but I don't see why it would be impossible. I actually saw one at a store (not for roof but anyway).

Great responses, thank you! How plausible do you think it is that perovskite cells will see commercial adoption in the next 5 years? This [1] makes it seem like it is possible we will see the first roll out in 2025 but with significant uncertainty about our ability to scale up mfg after that.

[1] https://spectrum.ieee.org/amp/perovskite-2667580324-26675803...

Can't but feel perovskite to be a fad no different than DSSCs before.

And I say it as someone who's researched them at EPFL in Michael Graetzel's laboratory.

I don't think the technology will ever be efficient and most importantly stable as needed.

Does it degrade when hermetically sealed? I thought it was down to moisture and getting it in a double glass would be good enough

Moisture heat and light all tend to degrade perovskite.

Perovskite is a family a materials by the way so many of these issues can be sortened out.

I'm overall just skeptic.

> ... heat and light all tend to degrade perovskite.

Have no knowledge in this field but if your solar panel material degrades in light that would seem to be an insurmountable problem to me.

For "tend to", finding a design that avoids it is surmounting it.

Or if you get it slow enough.

Output degrades at about ~1% a year. EOL for a solar panel is typically about 30 years. It will still generate power but if it is a commercial operation then it will likely be in consideration for replacement. I imagine there would be a large secondary market in the future for used but still functional solar panels as utilities and smaller solar farms replace the panels and possibly funneling into the residential market.

They don't have to be as efficient if they're 100x cheaper and lighter to boot. The stability is really the only issue.

At planetary scale it seems to be quite important not having to replace the whole fleet every few years don‘t you think? Just from a resource perspective this planet shouldn‘t drown in defunct solar panels.

Luckily, they are made of just a few basic, very recyclable chemical elements. Aluminium is fully recyclable. Glass is fully recyclable. Silicon gets purified as part of its manufacturing process. And that just leaves the plastic backing.

Just a thought, but recycling glass is extremely energy intensive.

https://www.nrel.gov/docs/legosti/old/5703.pdf

"Recycling of glass does not save much energy or valuable raw material and does not reduce air or water pollution significantly."

Yes, that's why I said the stability of perovskites is the only real issue.

Weight of the panels is not a significant portion of installation and labor costs.

Square footage (aka surface area) and installation surface challenges are.

Roof mounting is expensive. Supporting snow and wind loads is expensive.

Reducing dead weight is only going to help a tiny percent, as even if they weighed literally nothing it would not meaningfully change the load calculations.

That’s in a residential/urban context. I would assume (but have no actual idea, just common sense) that weight matters at least somewhat significantly in an industrial/grid scale context, where shipping and labor are for MW scale systems rather than kW scale systems. I’m not sure how important the weight is here, but it seems reasonable to think that it might be more significant than in a residential/urban context. Could also mean less land use, less structural steel, less maintenance for a system of the same capacity, etc.

It isn’t. A 500 watt panel is about 71 lbs, for 27.5 square feet. That’s the typical commercial panel. A little smaller than a normal 4x8 sheet of plywood.

Any structure designed to withstand 100 mph winds (typical in mild areas with no hurricanes or strong gusts) needs to be able to handle 25.6 psf - or 704 lbs - per panel just from wind load. Roughly 10x the panels weight.

In most of the US, add on snow loads from 20-100psf or more. I’ve installed panels in areas with 150psf design snow loads.

In the 150psf snow load area, that meant an additional 4125 lbs for that same 500 watt panel, each. Or about 58 times the weight of the panel. Steep angles (30 degree or more) can allow reducing that, which is a good idea.

So for instance in that area if not mounted very steeply, the racking needs to be able to support 71 lbs (panel) + 704 lbs (wind) + 4125 lbs (snow) per panel. Or 2.5 tons, give or take, for each 71 lb panel.

The panel is about 1.5% of the weight in that scenario.

And that is with no safety factor.

Now the roof has already been designed to bear these loads of course - but not as point loads randomly through the roof deck. So whatever anchoring/racking needs to transmit the forces effectively into the roof in a way it can handle without letting water through, and hopefully without making it impossible to maintain the roof either. And if in an area that freezes, without giving areas for ice to form and jack the roof/panels apart.

That isn’t trivial.

Great response, thanks for taking the time to write out the numbers!

What do you think about the implications for transportation, maintenance and land use? I have zero idea what the balance of those costs would be for a grid-scale solar farm, but ostensibly going from let’s say 20% to 21% efficiency means you need 5% less land, weight to transport from factory to site, fewer panels to inspect/build/install, fewer to purchase, etc.

I’m sure someone else has a better idea how much it would affect the LCOE than I do!

It isn’t going to make something economic that previously wasn’t. If the costs are sufficiently low (unlikely) it might have positive ROI in some scenarios.

Generally though, solar projects are go/no-go due to things like cost of money and electrical sales pricing agreements + site specific variables like insolation, flatness/road access, cost of local labor, local weather impacts on racking costs, access to transmission, and bulk wholesale costs of materials.

It’s hard to beat flat land out in the open desert near major urban areas with nearby highways and transmission lines, for instance.

What you’re talking about is likely at most half a percent of that equation.

Yep, that’s what I figured! And hence my original post at the top of this thread, suggesting that it’s easy to feel like we have sort of “solved” solar from a panel efficiency perspective and it’s everything else that we still need to improve on (grid infra, storage, etc etc), and additional percentage points of efficiency won’t really mitigate the existing limiting factors.

>That isn’t trivial.

Thank you. So often when discussing solar (or EVs) we see bizarre extrapolations of potential install rates that don't account for the fact that huge swaths of the country (the majority of places here in Canada) have real challenges with installation. These are not insurmountable, by any means. But those challenges are reflected in overall cost, making some of economics less favourable.

It’s always easy if we don’t know/talk about the hard parts!

In my experience, the panels themselves are at most 1/4 of the cost of any given system, even when discounting labor and permitting costs.

What proportion of the costs are down to permitting?

From 50% (or more) to 5% (or less) depending on local jurisdiction and scale.

If they want it to happen and aren’t greedy? It’s rarely a major problem. Otherwise, sky is the limit.

I know of a couple sizable projects that finally got cancelled because the local AHJ (authority having jurisdiction) finally just got too greedy. In one case they threw on an extra couple hundred grand worth of city park improvements as a requirement on a couple million dollar (small) project. Developer ended up walking away, as that was the fourth time they did that.

Some folks just can’t help but make it lose/lose.

So it sounds like large scale deployments have a larger proportion of the cost be due to permitting, and smaller scale (like on a house), would be fairly low by comparison.

Not really. It depends on how hard you can be squeezed, and how much they want to squeeze you.

Most large scale installations (if they’re smart) will be in areas where the planning authorities don’t have a lot of leverage.

Most residential installations (if they’re smart) will be in areas where it’s politically untenable to squeeze homeowners for outrageous fees.

Then you have the other places.

Either way, even large fees for a larger installation will be a small percentage of the total. $2000 worth of fees for a homeowner will be outrageous percentage wise.

For my residential 6.6kW ground mount array, less than 0.9%

But even if we reduce installation and labor costs (and as other comments have mentioned, weight may not be the biggest factor here), my understanding is that solar is already the cheapest energy source per kW by a pretty significant margin. The problem with getting it to replace carbon-based fuel sources is all the other issues GP comment mentioned re storage, grid, etc.

That said, it's not like one thing is dependent on the other, so good to see efficiency increasing regardless.

Interesting podcast discussing, among other things, why being cheap on cost/kwh isn’t enough

https://podcasts.apple.com/us/podcast/volts/id1548554104?i=1...

Volts is an awesome podcast for energy-transition enthusiasts.

It’s only cheap if the land is free, transmission costs are free, and your power needs are intermittent. None of those are true in reality.

The “solar is the cheapest form of energy” is a marketing gimmick. If it were true free market forces would already gone 100% solar because the purpose of energy companies is to make money.

You seem to have misread my comment, or just didn't read the whole thing, because I mostly agree (except for the part about "land is free" - most of the solar costs I've seen take into consideration full installation and financing costs). I said it was cheapest on a pure per kW basis. But exactly for these reasons e.g. the intermittent nature of it we can not currently go 100% solar without storage solutions, infrastructure improvements, etc.

Point being it is those areas that will need to be solved (i.e. come down in price) and that marginal improvements in panel efficiency aren't the limiting factor in going to 100% renewables.

> If it were true free market forces would already gone 100% solar

EVs might be cheaper than ICE cars now (including energy expense over their expected lifetimes)... But that doesn't mean the free market should be 100% EVs (all of a sudden).

While land isn't free, solar doesn't need land, it needs sun. By which I mean there's an awful lot of free rooftop space available, and a lot of space which would benefit from a roof (like car parks at the mall).

So it's not competing for land against say buildings or agriculture.

Now sure, the owner of the roof may want some rent etc, but that really doesn't alter the cost of the energy, it just spreads the benefit.

Why are we choosing rooftops instead of a central location for solar, like we do for all other types of electricity generation?

At first sight, it seems economies of scale would make it easier to have one company handle a solar power station, rather than now having to pay expensive home solar panel loans and maintain them.

We are choosing central locations. Google utility-scale solar.

transmission of electricity has losses.

And 2 solar panels only make 2x as much energy if exposure is the same. So go where the sun is already done if the sun is above your head.

>> Why are we choosing rooftops instead of a central location for solar, like we do for all other types of electricity generation?

solar is different to say coal, because the economics of home-generation, and grid generation are not miles apart. In other words it's not like I can have a coal-fired power station at home, but I can have solar panels. Up to now electricity generation has been constrained to large-scale (hydro, coal, nuclear etc). The advent of solar, and to a lesser extend small-scale wind and hydro, makes local generation more accessible.

>> rather than now having to pay expensive home solar panel loans and maintain them.

Solar maintenance is minimal. (Again, not like a turbine generator.) Loans are a function of capital. It takes capital to populate a home roof, and capital to build a power plant. If you have no capital then the point is moot. As an individual I have enough capital to fund my solar system without loans. (I get about a 14% return on that capital.) I don't have enough capital to fund a power station.

Other benefits of home generation include more resilience should the grid fail. So for example, after a storm, power lines may be down, but I get electricity during the day. That's a bonus though, not the main driver.

So to answer your question - it's not either or, it's both. There's a lot of roof-top solar in my city (measured in gigawatts), there's also solar farms generating power.

Lastly' I'll point out that distribution _from_ my house is cheaper than from a plant, because I generate a few spare kw, and the wire already coming into my hose is sufficient for that. So no new (grid) hardware is required.

> In other words it's not like I can have a coal-fired power station at home, but I can have solar panels.

Not that it matters to your point, but you can, they're just awful — there many reasons why everyone moved away from heating homes with open fireplaces.

(My current apartment in Berlin is old enough to have a chimney, but there's no unit attached to it; likewise the house I grew up in back in the UK has a chimney, but it was bricked off since before I could remember, possibly before I was born).

That’s just heat generation, not power. For power generation a larger generator will be more efficient than a small one.

For heating this does not matter.

Indeed, but that all adds to the reasons why, despite it being possible, nobody does it.

There's relatively little economy of scale in placing solar panels (compared to other energy sources). A lot of the scale advantage has already been achieved by mass producing identical panels.

Solar farms (on land) are far cheaper to install and maintain than rooftop.

Utility scale farm might be ~$1/Watt for installation. https://www.nrel.gov/solar/market-research-analysis/solar-in...

Residential is ~$3/Watt.

Commercial rooftop is somewhere between but still more expensive than a solar farm. Rooftop has the advantages that it delivers where the load is, and there are often subsidies available, plus some marketing kudos. But if just considering land cost by installing rooftop instead of solar farms is usually not an economic tradeoff.

And note that agrivoltaics (dual use solar+agriculture) is more expensive than pure solar farming: https://www.pv-magazine.com/2021/03/26/cost-comparison-betwe...

>> Residential is ~$3/Watt

Installation costs are going to depend a lot on your location, roof type and to an extent size.

My (residential) install (9600w) was around 60c per watt.

Ymmv

The Lazard report is pretty reliable, some 2023 figures for LCOE from: https://www.pv-magazine.com/2023/04/14/average-solar-lcoe-in...

  Utility-scale solar PV comes in anywhere from $24/MWh to $96/MWh

  Unsubsidized residential rooftop PV has an LCOE between $117/MWh and $282/MWh,

  the LCOE of community and commercial and industrial (C&I) solar ranges between $49/MWh and $185/MWh.

  When factoring in federal tax subsidies under the US Inflation Reduction Act, including domestic contest provisions, rooftop PV comes in at $74/MWh to $229/MWh, and community/C&I rooftop PV at $32/MWh to $155/MWh.

It is hard to find good comparative figures with similar assumptions (e.g. incentives, location, year), but the above give a ballpark.

The only people pushing rooftop solar are the ones who haven't thought about it very hard. Roofs are expensive. We build them to A) hold up the most likely loads and B) keep our water. Putting a bunch of solar panels on them makes them worse at both, especially when we have a lot of unused land. I mean monumental amounts.

I presume you mean "B) keep our water off"?

I guess the viability of roofs depend on the construction techniques in your area. In my experience roof's here have plenty of extra load availability, and I'm not sure what effect the panels would have on water.

In my area something like 5GW of solar has been installed on rooftops, and that has moved the needle.

And of course land availability varies a lot by country.

YMMV.

Geez yes I meant keep out water but I'm on mobile and not very careful! Yes, if we're talking about Hong Kong or somewhere with extremely high urbanization maybe the calculus is different, but that's a political issue more than a technical one. Every city on earth has land not utilized for living or heavy ag within a very reasonable distance.

If it's a marketing gimmick, they're fooling a lot large power companies:

"Solar and battery storage to make up 81% of new U.S. electric-generating capacity in 2024"

https://www.eia.gov/todayinenergy/detail.php?id=61424

They’re not fooling those companies. Those companies are happy to push “we’re green”.

Not saying it’s a bad thing, but I don’t buy that any company is being fooled by this.

Plus, they charge more for greener power.

Ah yeah the classic companies that forego profit to put a sticker on their logo

We are in the process of seeing a rapid and vast shift to solar.

Infrastructure takes time to roll out so the timeframe of solar is maybe 10-15 years before we see 50% of the worlds power switch to solar. If the 10-15 year estimate is true, this would be a breakneck speed for such a fundamental infrastructure change.

Do you throw out things you already own when you see that an alternative has become cheaper than what you paid?

> If it were true free market forces would already gone 100% solar because the purpose of energy companies is to make money.

I’m not going to speak on whether solar truly is the cheapest form of energy as I have no idea whether or not that is the case. But I’m going to suggest that one of your premises is wrong: we don’t have a true free market in energy — see all of the subsidies that gas, oil, and coal companies have gotten (for quite some time).

Yes, solar gets subsidies, too, but comparing a relative newcomer to entrenched players makes this a lot less clear of a picture than you’re painting.

https://en.m.wikipedia.org/wiki/Fossil_fuel_subsidies

Existing power generators have deprecation schedules measured in the decades. The free market forces are going to prevent those investments from becoming a write off today.

Seems a bit counterintuitive for panels made out of lead. Being toxic and not very long lasting doesn't seem like the best combination tbh.

Weight is relatively irrelevant, as long as people can lift them within health and safety regulations. For shipping in container ships or on trucks, at this order of density magnitude, it's ~free. Re. "Installation and labor" ... many issues here from a western context are likely safety/regulatory driven and not technically driven. In any event, weight reductions are meaningless if you aren't allowed to install it on the roof, or if the first gust of wind breaks it in half because it's acting like a sail. In many cases likely the mounting/frame weighs more than the panel.

Macro-demographically, with more people moving in to apartments globally, the residential roof thing is more a temporary western thing than a utility scale solution to global residential macro energy needs. The Chinese know this better than anyone.

The true economics of industrial processes are rarely clear to consumers.

Weight reductions? Really? Can you elaborate on why that would help?

Every PV installation I've seen uses heavy blocks of concrete to keep the panels from taking flight in the wind. The panels themselves are already very light.

Weight might still have a place such as states and countries with hurricanes, tornados, and other windy storms

In general I agree. But every efficiency gained will translate to less land used for solar, or a longer period of energy surplus every day (which means less deficit that has to be made up in other ways during the non-peak hours). Either of those things make me happy.

A manufacturing efficency would, this won't. The cheapest panel wins.

Not necessarily the cheapest, but the one with the best combination of efficiency, cost, durability, and possibly other properties I'm overlooking.

I meant in the marketplace as bought by actual consumers. So, the cheapest.

Isn’t GP effectively saying $/W? I assume that’s how most people evaluate price… “I want to install a system that will pay itself back in the smallest amount of time while staying below my max budget.”

That would be the prosumer. A consumer buys the thing with the lowest sticker price. See for example triple pane vs single pane windows.

I guess be careful how you use prosumer in this kind of thread, since it often might refer to a producer-consumer (ie someone who can get paid $$ to export energy back into the grid, as opposed to just saving $$ from avoided grid import).

I assume you meant professional consumer.

> See for example triple pane vs single pane windows.

Bad example but I understand your point. Bad example because it takes significantly more expertise to model the savings from triple paned window with low-e coating and vacuum sealing vs double paned vs single paned (ie it requires a proper building energy model in eg EnergyPlus). Also bad because the triple paned window might be immediately disqualified by the budget constraint above. In any case, I think you are just saying people are likely to pick the thing which has the least friction, whether that friction comes from cost, installation challenges, etc etc.

And yes, sometimes upfront cost is the most significant thing which affects people’s willingness to adopt a certain home energy retrofit but payback period does play a big role in people’s decision making, as well as their ability to get funding for it (government rebates), or they may simply be going with an installer who pays them to install it and they certainly care about payback period!

> Bad example because it takes significantly more expertise to model the savings from triple paned window with low-e coating and vacuum sealing vs double paned vs single paned (ie it requires a proper building energy model in eg EnergyPlus). Also bad because the triple paned window might be immediately disqualified by the budget constraint above.

I'm confused by this as that is exactly the point made. :)

I was watching a NASA iTech talk if memory serves about vacuum windows. He opened with an overview of the market and adoption trends and was quite flustered. Small market, little in the way of R&D and many challenges in manufacturing them in the US.

One of the things that flustered him is being told by sales reps that those types of windows are "pointless and not worth it" and so on. Digging deeper the reason he was steered away from them is simply because they were up not as lucrative to sell in that moment at time for that provider. I can try and dig that up if you want as I recall this anecdote vividly and my thinking diverged from yours in that moment.

Turns out most people have no idea what any of it means you see? So indeed friction, perverse incentives, financing, lack of consumer education and so on. Some terribly sad amount of people still opt for single pane instead of double which shouldn't make any kind of sense.

Remember also that people often move houses. Anyway the upfront cost ended up dominating rather than the payback period. Folks aren't that rational or liquid.

Out of curiosity try modeling it out and pitching it as a choice if you know anybody looking;

Chinese panels <20% efficiency, cost X, payback period Y vs Unobtanium panels 25% efficient, Cost X+, Y++.

But don't try to steer them or mention the payback period unless asked. It isn't as straightforward as we would hope.

> I'm confused by this as that is exactly the point made. :)

Sorry, I was trying to indicate that it is significantly simpler for a random uninterested or mildly-interested consumer to evaluate the cost-effectiveness of solar than it is windows. Most people have a good sense of how much their energy bill costs, and there are plenty of cheap and free tools which let you accurately estimate how much energy you will save from a PV system, but even basic mental math is enough. As opposed to windows, which are significantly more complicated because they involve thermodynamic modeling of your home.

You should read recent work by Zachary Berzolla, who is now working in Maryland’s department of energy on their commercial buildings decarbonization program. His MIT PhD dissertation is specifically about willigness-to-pay for home energy retrofits (but especially heat pumps). Not sure if it has been posted to DSpace yet but I believe some various conference/journal papers are online already.

Unfortunately, I think a lot of people don’t really know how much their power costs. They just pay the bill they are sent, maybe even on autopay and they don’t even look at the statements when they come in.

So, they don’t know how much their power costs, and they don’t have a clue how much solar would cost to buy, install, run, etc… and how that compares to the payback period over time, etc….

These are the people that I think we need to reach.

> They just pay the bill they are sent, maybe even on autopay and they don’t even look at the statements when they come in.

On the other hand there are a significant number of people who are energy-burdened, and they often have the most inefficient homes (leaky or non-existent sealing, little insulation, old appliances, etc). These homes often have the least ability to take advantage of government rebates since they may only be tax rebates while the retrofit still requires up the upfront cost to be paid. At the same time, high-income homes, though often much more efficient, also often use the most energy (since floor area tends to grow with home owner income, and space conditioning/electric requirements tend to grow with floor area). It’s easy to design incentive programs which have a big carbon impact but a bad equity impact in that they just end up giving money to people who would already be upgrading their homes without the rebates.

The person who knows just how much money they are spending on their bill every month will likely value the savings much more (ie the utility value of the savings are much higher), but they may also have far less awareness of the kinds of programs available to them for retrofitting their home or installing PV.

It’s a complicated problem, figuring out who to reach and how to drive adoption while balancing decarbonization and equity!

Who is talking about consumer panels?

no, more efficient solar panels at a given price mean cheaper energy, which means it becomes economical to use more energy, until you run out of ways to use energy usefully. usually the increase in energy demand is more than enough to increase the use of inputs, a fact known as the 'jevons paradox'

> a fact

Jevons Paradox is conjecture, not immutable law. Consumption based energy usage has been on the trend downwards in western countries for some time. Much of this is explained by efficiency gains (e.g., LED lighting, OLED screens, smaller chips, heat pumps, etc.)

https://ourworldindata.org/grapher/consumption-based-energy-...

it's just an observation that has held true for two centuries. pointing at consumption-based energy usage in a cherry-picked group of countries is special pleading; neither worldwide consumption-based energy usage, nor overall energy usage even in that cherry-picked group of countries, is falling. furthermore, the link doesn't show that even that cherry-picked measure is falling

the jevons paradox certainly isn't an immutable law

> cherry-picked group of countries is special pleading

It's hard to disentangle the rise of wealth in the world from increased consumption caused by cheaper energy. Certainly in the west capacity has become larger and...

> the link doesn't show that even that cherry-picked measure is falling

If you look again, they absolutely are.

Peak vs now in Mwh: US: 105 (2005) vs 97 Norway: 93 (2008) vs 78 Sweden: 69 (2001) vs 57 Australia: 80 (2011) vs 55 Germany: 63 (2006) vs 59 France: 61 (2005) vs 54

I could manually do this for the rest, but I shouldn't need to. You can see it for yourself. Here's the chart as line graphs (this option isn't obvious in the UI)

https://ourworldindata.org/grapher/consumption-based-energy-...

The only outlier I see here, in terms of developed countries, is South Korea, but that's perhaps as the country is still unevenly developed. In terms of developing countries, we of course see the uptake of HVAC, refrigeration, vehicles and other automation appliances to match the living standards of the developing world. But crucially, developed countries are showing that at the highest living standards, energy consumption is coming down.

> Peak vs now in Mwh: US: 105 (2005) vs 97

that's the chart i was looking at. you're cherry-picking years as well to get that answer. looking at all the years instead, the usa starts out at 94, peaks at 104, dips down to 87, and then rises back up to 97, with lots of wiggles in between. that isn't a picture of a secular decline in energy use, it's a picture of random fluctuation around an average. the others are largely similar. the main outlier among 'developed' countries is not south korea but the people's republic of china, though slovenia deserves a mention for falling 40%

why? what holds the usa line so steady? probably partly the usa has lost the ability to build not only high-speed trains, subway tunnels, highways, and leading semiconductor fabs, but also electrical infrastructure, but two thirds of energy use is non-electric, and other countries are succeeding in building trains and whatnot

the price of energy has not been holding steady during this time but rather going up, although in the last decade that is starting to reverse. perhaps the reason consumption-oriented energy use has gone slightly up rather than way down is the efficiency improvements you mention

[the following paragraph is completely wrong, and i appreciate the correction from h0l0cube. therefore so is my claim that the measure itself is cherry-picked. i had just misunderstood it]

this measure, incidentally, doesn't count energy used for exports at all. so if the us and japan are producing an energy-intensive good independently at the beginning of the chart, and then in the middle the us starts importing it from japan, the usa's line will go down while japan's remains steady, because japan's added energy use is for export rather than for consumption. that's what i mean about it being a cherry-picked measure even for those cherry-picked countries

> this measure, incidentally, doesn't count energy used for exports at all

Literally, it does. It's consumption-based energy usage. It's trade adjusted

> the usa starts out at 94, peaks at 104, dips down to 87, and then rises back up to 97, with lots of wiggles in between. that isn't a picture of a secular decline in energy use, it's a picture of random fluctuation around an average

The trend line is at best a flat line. Here's a picture of energy usage per capita over a longer timespan (inc. prior to the 80s when offshoring of manufacturing and cheap shipping came into play)

https://ourworldindata.org/grapher/per-capita-energy-use?tab...

> the main outlier among 'developed' countries is not south korea but the people's republic of china

As I mentioned this is due to it's growing wealth, not the greater economy of power. Even though China has been growing phenomenally in GDP, it's GDP per capita vs consumption-based energy usage is going down (or yet to exceed 2011 levels):

https://ourworldindata.org/grapher/change-energy-gdp-per-cap...

> Literally, it does. It's consumption-based energy usage. It's trade adjusted

thank you for the correction; i had misunderstood what the trade adjustment was

> The trend line is at best a flat line

i think that is an excellent description of it, as well as of the expanded and less debatable chart you link here, at least over the last 50 years, since the energy crisis began. before that we were seeing a much different trend

i agree about wealth being the primary driver of energy use. but that's precisely the story jevons tells: you improve your steam engine to use less coal, so now you can build a railroad, which is a form of wealth. previously a tonne of coal cost £100, say, and produced 30 megajoules of work, which had a value of £200 in pumping out a mine or £50 hauling goods on the railroad. with your new engine you improve efficiency to 0.3,% and get 100 megajoules, so pumping out the mine now costs £30 per £200 produced, but now the railroad is viable because it produces £50 - £30 = £20 net. the railroad consumes much more coal than the mine did, so you're using more energy more efficiently at the level of the machine, raising your gdp, but producing less value per tonne of coal

well, i got the numbers a bit wrong, but hopefully you can see what i mean

> i agree about wealth being the primary driver of energy use

That's not what I meant. I was referring to the development of populations. Populations that didn't have what the global north would call 'the basics', electricity, household water/sanitation, lighting, adequate heating/cooling, refrigeration, as well as 'basic luxuries' like television/computing/internet and personal transport, increasingly now have access to these technologies (even in 'underdeveloped' countries). This is going to increase energy (not just electricity) usage. But once 'the basics' have been met, they will already be in line with the efficiency standards adopted by the west (maybe even moreso because energy is more expensive in the global south)

So, a counterargument might be that AI will become part of our 'basic' lifestyle, and that we will see a resurgence of demand again, but we're seeing that cloud based compute is acceptable for the vast majority. So even though, say, internet search was energy intensive at its inception, it has largely been amortized and itself energy optimized to not really raise the bar of energy usage. Custom silicon, tighter semiconductor nodes, newer algorithms, photonics, and eventually yet-to-be-discovered technologies like room temperature super conductors can again bring the energy usage down for compute, increasing efficiency not just for AI, but across the board.

that's what i'm talking about too, but yeah, i don't agree that there's a finite set of 'basics'; i think it's dependent on how much you can afford. for 50 years that's changed only very slowly in rich countries, and now it's about to change faster than it did 200 years ago, not because of the kinds of slow efficiency improvement at the point of energy consumption, but because of photovoltaic panels

things that already exist but could get much cheaper due to cheaper energy might include ai, as you suggest, but also personal computers, oocyte cryopreservation, ecm machining, space travel, weekly air travel, personal helicopters, atmospheric carbon capture, caribbean cruises, making things out of aluminum or titanium instead of plastic, cnc machining rather than casting or stamping, cars, buildings, photovoltaic panels themselves, etc. and presumably there are other things that haven't been invented yet because they'd be uneconomic

Energy use goes down because Cost of energy keeps climbing.

Not so much cost of energy, but the price of energy charged to consumers. As ever, the cost + margin sets the minimum for the price, but the price maxes out at what the consumer is willing to bear. Though to be fair, in the last 5 or so years, the cost of energy has risen due to the need for a dynamic network, entailing more and bigger transmission lines, but that doesn't explain the reductions entirely.

Anyone who remembers how pervasive CRTs (later rear projection, and plasma TVs) and incandescent bulbs were, can sense that things are far more efficient around the home. Better standards for HVAC (using heat pumps vs heating elements) are a big deal. A big driver of this has been large buildings where the cost reductions from efficiency can mean millions saved. Business errs towards efficiency. Another driver is that the more efficient stuff also seems to just be better too (like OLED screens)

We also have chips that run on about the same energy (or even less) and provide far more compute than 2 decades ago. Looking further back the ENIAC used 174kw of power for 0.005 MIPS (5000 additions). By comparison an M1 has 2.6 TFLOPS (2.6 trilling floating point operations) for 40-100 watts. My cores are mostly idling. In fact most of my computer usage happens on my phone which runs on far less energy still.

Then consider energy for transport. In the last 5 years, telecommuting has become so normal that fewer people are commuting to work, particularly cities.

note that most energy sold isn't electric. jevons made his observation when none was

not all consumer energy prices are rising, and the situation you describe where the producer captures the whole consumer surplus only happens in the absence of competition

jevons says people will compensate for more efficient tvs by buying more and bigger tvs, and for more efficient hvac by building bigger houses, living in hotter areas, and installing air conditioning in more spaces. this has been a major trend of the last 25 years you're talking about, in fact

similarly, it's true that the amount of energy per flop is going down, though not nearly as much as you might think; on the cpu it was about 2000 picojoules 30 years ago, and closer to 500 picojoules today. (the numbers you give for the m1 would work out to 16 picojoules, but the real numbers are even better. 2.6 teraflops is the graphics card, which uses 11.5 watts, which is 4 picojoules per flop. most computers lag far, far behind that in efficiency.) but the amount of energy spent on computers keeps going up and up

> producer captures the whole consumer surplus only happens in the absence of competition

Collusion is the norm. Especially with electricity suppliers where they jack up the rates because most people won't be bothered to check for better deals and switch.

> jevons says people will compensate for more efficient tvs by buying more and bigger tvs

Doesn't matter if the efficiency gains outcompete the consumption increase (which is my whole point):

> As shown in the graph below, between 2006 and 2012 the average energy consumption of televisions dropped by 57%, while at the same time average screen size increased by 30% and average price decreased by almost two-thirds.

https://appliance-standards.org/blog/why-recent-progress-tel...

> for more efficient hvac by building bigger houses, living in hotter areas, and installing air conditioning in more spaces. this has been a major trend of the last 25 years you're talking about, in fact

McMansions became a thing leading up to the 2008 financial crisis, and was a feature of cheap finance, not energy cost. Most people want a house that's affordable, close to amenities and doesn't take an army of servants to clean – so a house in the suburbs and within distance of the city. The factors here are land price and construction costs.

> but the amount of energy spent on computers keeps going up and up

Might be true, but it would be nice to see some data to back that claim.

collusion fortunately doesn't allow exxon to bar the importation of chinese solar panels (except in the usa) or their sale to consumers, and in most of the world, electric utility regulators force utilities to pass on some of the savings from utility-scale solar to customers. even where they don't, heavy industry can negotiate contracts, because electric utilities in paris don't collude adequately with those in bavaria, kansai, and virginia

a 130% increase in tvs per person would cancel out the 57% reduction you cite in power per tv, and presumably if you measure over any other period of time, the energy consumption drop won't be that large, because that was the crt–lcd transition, but screens had been steadily getting bigger for decades and have continued to do so. if you include computer monitors with tvs, it seems likely that the increase over the last 25 years is a lot more than 130%. certainly it is in public places (dentists' offices, intercity buses, airports, restaurants)

people have been building continuously bigger houses for centuries; it's not a phenomenon limited to zero interest rate mcmansions. suburban sprawl is one well-known manifestation of it as for computers, unfortunately i don't have the data here at the moment. you see handwringing about it from time to time. i'd like to see it too

> 130% increase in tvs per person

Is each person watching all these TVs at once Matrix style?

Even if you've got your laptop, iPhone and iPad all unlocked and playing videos while watching your TV, it's not really adding 30% of energy use compared to the 50 inch TV.

> certainly it is in public places (dentists' offices, intercity buses, airports, restaurants)

TVs in airports, bars, diners, buses, and dentists offices have been around since the 80s. The difference now is that all their CRTs are replaced — a clear energy drop. (Source: own experience, also you can watch movies from the time)

Lit billboards have been around for a long time, but instead of massive lamps, they are just LED screens. Tokyo used to be all fluorescent bulbs, now LED screens again. It might not be like for like, could maybe be an energy increase, but public-space advertising is the best argument you've got here, but I don't buy that it's a huge driver of energy growth.

> you see handwringing about it from time to time. i'd like to see it too

I've provided plenty of data to back myself, and to suggest otherwise is disingenuous. Given the amount of claims you're making without any supporting evidence, I can only think your argument is based on conviction, not fact. I'll let you have the last word here.

oh, it's quite common for different people to have different tvs on at the same time in different rooms of the same house, when previously they would have watched the same one. or for a tv to be left on displaying a screen saver. a backlit lcd uses the same amount of power whether it's showing black, blue, a static desktop, random noise, or video

i've seen many more tvs in public places in recent decades, often replacing things like flip-dot displays or printed menuboards. mcdonalds now displays their menus on a wall of tvs instead of printed on plastic, for example

it's disappointing that you've chosen to attack my integrity, but it seems to be based on a misunderstanding. i have not claimed that you have not provided data. on the contrary, i have found your data highly informative and educational. i only meant that neither of us has convincing data about total computer power consumption, and that i would like to

you'd think, but energy use is going up, not down. see https://news.ycombinator.com/item?id=40700537

the price of energy has started to fall again in the last ten years

Solar cells that can be printed like a fabric or rolls of paper, solar cells that can be spray painted, solar cells that can be grown from microbes (fastest and cheapest!) are technologies that will make current solar technology look primitive.

That may take a while, but immediately though, cost-effective solar shingles would be so much better than wasting material/labor of 3 layers. Tesla roof and GAF Timberline have products, looks like GAF costs the same as other materials with tax credits. But if this can get cheaper, its a gamechanger.

How would a spray-on material have macroscopic structures like bus bars?

https://www.solarreviews.com/blog/solar-paint-hydrogen-quant...

Can you link me to the "solar cells grown from microbes" please? Not heard that one

https://www.sciencedaily.com/releases/2018/07/180705084215.h...

you'd think, but the roll-to-roll processes of 15 years ago somehow turned out to be more expensive than just slicing silicon thinner, and they've been driven out of the mainstream market

https://www.amazon.com/Xunlight-XRD36-Flexible-Rollable-amor...

This is 300W, works out to be 0.75 cents/Watt, which seems decent? Also, they show videos of heavy equipment driving on it. Which means they can be laid flat on the ground, with very little support structure. They can be laid out on parking lots. There must be reasons why this isn't widespread yet. I wonder what those are.

(edit) This post says panels are 15 - 20c/Watt: https://news.ycombinator.com/item?id=40703758

0.75 cents per watt would be two dollars and twenty-five cents for 300 watts. this costs 225 dollars, 100× that much, and almost 10× the price of conventional solar cells in the international wholesale market. you can see those are 8 cents a watt in the thread you linked, less than half the price of a year ago

solar cells that can survive being driven over do exist, but they only survive a few months of being driven over, which is why they aren't widespread. a much more sensible way to put solar panels in parking lots is to mount them above head height, which is relatively cheap in non-snowy areas. this has four big benefits over putting them on the ground:

- they last 60 years instead of 60 days

- they produce power even when cars are in the lot

- they provide shade to people and cars

- they can be angled toward the sun, increasing yield

hope this is helpful!

Thanks for the insights! Typo on my previous post, its 75 cents/watt vs 10 - 20 cents/watt. Too bad they can't survive more than a few months!

Percentages can be funny. 2 percent higher efficiency is 10 percent more output if you go from 20-22.

If the costs aren’t too egregious that can affect the break even point quite a bit.

And the situation with embodied carbon footprint. Which we pointedly do not talk about.

That would be two percentage points. Two percent, I would expect to be from 20 to 20.4%.

Technically correct but not colloquial usage. GP got that scale right, except they've actually managed going from around 20% to around 25%.

Solar would better be considered solved when the companies installing it don’t require large loans, liens on your home, and door to door salespeople. It’d be nice if having it installed was more similar to having a plumber or regular electrician stop by

I guess I meant from a technical efficiency perspective, rather than the social infrastructure around it. It’s unclear to me (and frankly seems unlikely?) that few more percentage points will fix the very real structural issues you raised!

At the same time, even if we do solve those issues, and we get solar panels on the roof of every home, there are still significant challenges to overcome as it’s unlikely the average home can become fully energy independent (especially if there is electrified winter heating), and anyways, the energy demand from single-family residential housing is only one slice of the overall energy pie. In any case, it would certainly help if we did that!

That's because putting solar on rooftops makes little economic sense. The overhead in terms of installation cost was relatively insignificant when PV panels cost 20 times as much - as they did 15 years ago. These days, the cost is swamped by installation costs.

These days, the only place it makes sense to put PV is flat on unused or unproductive land. The 95% drop in PV panels means that it is no longer economic to bother with mechanical tracking in solar farms. Integrating panels into a domestic roof destroys the incredible cost advantages PV has over alternative power sources.

I just installed 7.3kW on my roof, for a total out of pocket cost of $8000. (after rebates and free gov money)

It looks like I'll make about $1000 of power per year, so I have an 8 year payback, after which I'll have free power for another ~20 years.

How on earth does that make "little economic sense"?

I think they probably meant from the perspective of society writ large. If you you are choosing between spending $10m on PV with the primary goal of decarbonization, are you better served by spending it on subsidizing rooftop PV or on utility scale PV? Probably the latter.

> “In a base comparison, without considering subsidies, fuel prices, or carbon pricing, utility-scale solar and wind have the lowest LCOE of all sources. Utility-scale solar PV comes in anywhere from $24/MWh to $96/MWh, while onshore wind registers the lowest possible LCOE over the shortest range, from $24/MWh to $75/MWh. Offshore wind’s LCOE ranges between $72/MWh and $140/MWh. … Unsubsidized residential rooftop PV has an LCOE between $117/MWh and $282/MWh, while the LCOE of community and commercial and industrial (C&I) solar ranges between $49/MWh and $185/MWh. When factoring in federal tax subsidies under the US Inflation Reduction Act, including domestic contest provisions, rooftop PV comes in at $74/MWh to $229/MWh, and community/C&I rooftop PV at $32/MWh to $155/MWh.” [1]

[1] https://www.pv-magazine.com/2023/04/14/average-solar-lcoe-in....

People often do this calculation without considering the time value of money. If you took that same $8K you spent on your system and put it in a bank CD; you would earn about $420 each year in interest. That means your real savings for the solar system is only $580 a year, not $1000. That makes the payoff time expand from 8 years to nearly 14. Still might be worth it to you, but also might not.

True, but consider grid electricity prices increasing over those decades too.

they won't increase; they'll drop like a stone

Huh, my city has already approved rate increases for the next 5 years! (it has increased every year forever...)

yeah, it'll take a while for the new technology to trickle down to the more backward regions

I find your comment misleadingly pessimistic. It makes more sense to compare ROIs directly (or equivalently, payoff periods) rather than subtracting them and finding a residual payoff period. Just annualize everything (depreciation, inflation) and see which one has a higher ROI.

1. You need to take into account depreciation of the value of the panels. They degrade in performance and eventually will be worthless after about 30 years.

2. You need to take into account inflation against the CD roi (or conversely the /appreciating/ value of the dollar value of the energy produced by the panels). The post-inflation value of the bank CD is going to be about 2% per year. Inflation does not need to be corrected for the solar power option because it produces energy instead of dollars.

(1000/y-8000/30y)/$8000 = 733/8000 = 9.1% depreciation-adjusted ROI from solar panels

5.5% - 3.3% inflation = 2.2% inflation-adjusted ROI from bank CDs.

So solar panels are about a 4x better investment than bank CDs, contrary to your comment where they are somewhat comparable.

they won't be worthless, they'll be producing about 25% less power, and after that degrade very slowly indeed. but that doesn't matter much because 30 years is basically forever at any reasonable discount rate

bank cds do not pay a reasonable discount rate, it's true, but there are investments that do. maybe a nice index fund balanced with a money market fund?

you should also take into account the precipitous drop in electricity prices starting 10 years from now

You’re born short power, just like you’re born short housing (assuming no inheritance). Buying panels and buying a house makes you net zero on your exposure to power and housing prices. It is not fair to compare a hedge (solar) which reduces your exposure to risk (covering a short position on power) to a risky investment like stocks, which does not cover a position and increases your correlated risk.

Just comparing expected value is fine as a stopping point in your thought process if you are risk neutral — in that case, you should buy leveraged stock funds to maximize your expected value.

If you are like most people and assign some internal cost to risk, then covering your innate short position on power while also getting 9% return on investment after inflation is a no-brainer.

agreed, but it's probably less than 9%, because energy will get much cheaper

> real savings for the solar system is only $580 a year

> Still might be worth it to you, but also might not.

Are you trying to be intentionally obtuse?

With your numbers you're talking about putting $580 a year into my bank account for 14 years, and then me having free electricity for at a minimum another decade.

In what possible world could that be "not worth it" ?

did you do your roof at the same time? theres a big cost of taking the solar off and putting it back on when doing roof work which can kill cost efficiency if not lined up with the lifecycle of your roof.

Tin roof, it will outlast the rest of the house

the funniest part is where we assume everything stays the way it is. Centralized systems are more efficient but have poor reliability. What if they just stop supplying power? What if a kwh costs $10?

If we're counting overall RoI then the "rebates and free gov money" should be added in.

because you're paying $1.10 per peak watt in a place with a suboptimal capacity factor, and if we add the rebates and 'free' gov money, probably closer to $2. if we're talking about us dollars, low-cost panels cost $0.08 per peak watt (though predatory usa import tariffs double that) and we can probably expect a whole solar farm built with them to cost about $0.50 per peak watt, maybe $1 if you're talking about the us. that solar farm can be located in a place with a 20% or even 30% capacity factor, because it gets twice as much sun as your house, so it generates twice as much energy with the same amount of panels

so your utility company is probably going to make four times as much energy per dollar invested in solar panels as you are, unless you're in the usa, so they can sell it to you much cheaper than you can make it yourself

so probably if you'd put the $16000 or whatever into the stock market it would yield more than enough to pay your electric bill for those 30 years or actually forever

the panels won't wear out in 30 years either, though, and the reduction in risk may be worth it to you

> so your utility company is probably going to make four times as much energy per dollar invested in solar panels as you are, unless you're in the usa, so they can sell it to you much cheaper than you can make it yourself

I agree they can make it cheaper than me, but I don't think your second conclusion follows.

What they will do (and ARE doing) is simply increase their profit.

My utility company has already approved rate increases for the next 5 years (7-12% per year), and it has increased every year for the previous 10+.

So for me, the cheapest way to get electricity is to make it myself from my own roof. I made 933kWh in May for a bill of -$56.

yeah, to me it seems like a pretty reasonable investment even when you aren't being ground under the boot of kleptocratic government corruption, but an obviously good one when you are, unless they start levying a special solar panel tax ot something

Plumbers and electricians will put liens on your home too

> At the same time though, it’s starting to feel to me, to some extent, like we have kind of solved solar?

There are many metrics that affect power generation (or anything really). A few are:

- Power generation per unit area

- Power generation per unit mass

- Power generation per dollar

- Lifetime

- Decline rate (ie does the cell get less effective over time?)

- Flexibility (eg can you wrap it around a cylinder)

- Minimum size

- Maximum size

- Cost to repair

Where a given solar panel fits in the above vector space will change its applications. In some cases, size is paramount. In others, cost is paramount. Sometimes you need long-lived panels. A good example if solar panels for space probes. These need to be generate as much power for as little weight and it doesn't even really matter what the cost is. Also, making such a panel last 30 years might be irrelevant if the lifetime of the mission is 5-10 years.

So no, I wouldn't call solar "solved".

Fair points! I tried to clarify in my edit, but I was mostly speaking from the perspective of grid decarbonization, as opposed to more niche (to me) applications like space probes, because that is what my personal biases are towards. It seems like a few more percentage points of efficiency at the same cost/space (so ignoring something like perk pv) doesn’t meaningfully move the needle on decarbonization efforts any more and that other challenges dominate (grid infra, batteries, etc). And I actually mean “solved” in a celebratory manner - it’s awesome that we are at this point! But yes, I was totally overlooking use cases besides residential and grid-scale PV.

The better the solar gets, the bigger the incentive to restructure the surrounding infrastructure

I wonder if someone has done a science to compare the carbon impact of batteries and peaker plants. It might be that a lesser of evils fossil fuel is the solution if it's a sometimes thing and not the main thing. It also has the virtue of being supplyable in a renewable way with landfill gas recovery.

A battery made using the dirtiest grid that produces them at scale (110kg CO2/kWh of battery) breaks even compared to gas after ~200 cycles.

LFP batteries commonly used as stationary storage regularly do 2000 cycles.

Makes more sense to use that gas to make more batteries.

Peaker plants are considerably less efficient than base load gas energy, because the constantly running plants use a combined cycle. In both types of plants, the fuel is burned in a gas turbine to directly generate power; however, in the combined-cycle plant the waste heat from this is used to generate steam to run steam turbines, capturing additional power; this actually generates over twice as much useable energy per unit of fuel. Peaker plants cannot effectively use this mechanism as it has a much longer start-up time.

Combined cycle is a major reason that gas power plants are so attractive; the inability to use it in Peakers is a reason why they are so unattractive.

> Peaker plants cannot effectively use this mechanism as it has a much longer start-up time.

How much longer? If a plant is designed with a big priority to getting secondary generation up to speed quickly, how many hours will it need to warm up?

A lot of it depends on the exact type of usage, does it not?

Currently a lot of peaker plants operate a bit like "A power line failed, we need extra power NOW!" They get essentially zero warning and are expected to be at full power within 30 minutes. Dealing with that obviously leads to some issues, but in 2024 we could also fill that niche with battery storage.

When it comes to the energy transition, it's a bit of a different problem. We can reasonably predict weather, so the rough output of solar and wind is known several days in advance. If the forecast is predicting an overcast day with zero wind, any "peaker" plant will have tens of hours to warm up. Combine that with minor changes to reduce startup time[0], and it seems far less of a hurdle to overcome.

[0]: https://etn.global/wp-content/uploads/2018/09/STARTUP-TIME-R...

Completely agree. I'm sure there's still a lot of advances left in solar panels themselves, but we really do need to solve the grid, solve storage, essentially solve the base load

We won't be able to transition fully off of fossil fuels until we do

Not just grid scale. It can be the difference of panels on your roof generating enough electricity to barely meet your needs to having so much excess you can not only supply your home needs, but also charge multiple EVs at home.

I feel like that's a uniquely North American need. Most countries don't have the single family / multi car ownership structure. Still a real need. But implications of a country where large% of people can meet their own energy needs is interesting.

The implication would likely be that the share of population that can't afford it won't be able to pay for the grid alone either.

We have examples of this kind of situation in poor countries where the grid wasn't developed in the first place, and rich people use generators.

This still doesn’t address the duck curve aspect, or overnight usage. It is fundamentally impossible to use PV to directly power your wonderful super efficient heat pump to warm your home at night. You will still need batteries!

And? You can buy a rack of lifepo4 batteries for less than $10k and it will be enough power to supply you running your A/C or heat the entire night.

Well, my original point at the top of this thread was specifically about how battery mfg/infra/costs/dispatch (especially at grid scale) and so on seem like the limiting factors/primary challenges still being worked on, not PV efficiency, so I think we are agreeing?

There are heat accumulators.

Yep, those are just forms of batteries though so I think my point stands? There is lots of cool stuff you can do with PV+TES without even needing true thermal batteries, just using smart electric hot water heaters and dispatch, and even cooler things if you set up a peer-to-peer network of them! But again, that just goes to the point that other challenges have overtaken solar efficiency in urgency (which is a great thing to reflect on!)

Excess panels completely address the duck curve.

How? There’s a significant portion of the world/year where duck curve evening peak is after sunset and PV alone is never feasible and so batteries/wind/etc are necessary.

I was considering "duck curve" and "overnight" to be separate categories. Were you not? Then I will rephrase to say that excess panels solve the part of the duck curve that does not overlap "overnight", so the problem is reduced to just "overnight".

I do consider them separate, but the duck curve peaking is in the 6-8pm range typically before falling off to overnight baseload. 6-8pm is frequently after sunset. That’s what excess panels does not help, and is still separate from what is typically meant by “overnight” in my experience.

grid-scale batteries have already almost entirely eliminated the duck curve in California, and the costs of batteries continue to fall

Is it supposed to? Says who?

Incentivize car chargers at every workplace. Tap in with your personal card, and it could even be directly combined with your residential power bill so you're essentially charging using your own solar at a distance.

It’s still too large of a surface area for anything other than large on-ground installations. When you get to potential use cases like transport it still matters a lot how efficient the panels are. Let’s say running a cargo ship with on board panels for example

You’ll never be able to run a cargo ship effectively even with 100% efficient panels. [https://transportgeography.org/contents/chapter4/transportat...]

Maximum insolation is about 1kw/square meter. Assuming a very favorable TOE/kwh equivalent (11.6 mwh/toe) and 150-225 tons of fuel oil a day, we’re talking energy consumption of 1.7-2.6 gigawatt hrs per day to power a container ships main propulsion.

Assuming very favorable 10 hr insolation times, and 25% solar conversion efficiency, you’d need something like 680000 square meters - or 168 acres - worth of panels to even come close. Even with 100% efficient panels, it would be over 43 acres worth.

A Panamax container ship is only 2.3 acres in size, which equates to 9.5 megawatts of solar insolation peak.

So you’d need between 20-73 times more surface area, and very favorable conditions.

Not to mention batteries to smooth all that out. And a chance of storms. Or not having panels aligned perfectly.

Fossil fuels are incredibly energy dense, and these ships have to burn insane amounts of them already in very efficient ways to do what they do.

We've solved energy in middle of sunny, summer day. We haven't solved it in middle of winter night.

We're in the process, and may arguably already have, for the tropics.

Get more panels, maybe some batteries.

in antarctica you may need nuclear or ammonia storage or something, but the rest of us can just use batteries

Winter being a problem really depends on location. Seasonal variability is much lower near the equator. Also batteries are becoming a solved tech. Also wind is anti correlated with solar, it's stronger in winter and at night. So you want 50-50 to minimize the need for storage.

If you can reduce even 50% of the consumption from the grid that would be huge.

Luckily most people are sleeping in the middle of the night, so electricity usage is already quite low. That's why many areas already have a separate discounted Night Tariff.

Additionally, the winter might not have a lot of sun, but it usually does have quite a bit of wind. Build a combination of solar and wind power, and you've solved the biggest issue. The rest can be picked up by hydro and gas peaker plants (short-term), or battery storage and other new technologies (long-term).

the winter has plenty of sun and few clouds.

Irradiance of a surface is lower at the same time of day (think of shining a flashlight perpendicular to the ground vs at an angle) and the number of hours the sun is above the horizon is also less. Lower power at any given time of day and for fewer hours of the day.

Here’s a figure to demonstrate that:

https://www.researchgate.net/figure/Monthly-output-from-sola...

Any basic irradiance analysis of a surface in any day lighting simulation software will clearly demonstrate this.

Additionally, in winter, the peak energy usage will often be well past sunset in many locations around the world (and it will get even more pronounced with heating electrification). This is why batteries+wind will be very important.

I was very surprised the difference is that small. Intuition would suggest there to be a lot of sun in the summer given how hot it gets.

The day is roughly half as long but you get 40-50% per month ???

Really a lot more than I imagined.

Not sure if that involves changing the angle.(which would complicate things)

It isn't a popular or sensible technology atm but sterling or other heat engines do work and they work even better if there is a cold source. There are experiments with heating water with solar voltaic. I read they are reaching 70% energy conversion.

It seems there is lots of room for further tinkering. 40% isn't bad tho

It depends where. In Poland, we had whole 7 hours of sun this December.

Reply on news.ycombinator.com