This is very exciting! My understanding[1] is that silicon panels have largely hit their max efficiency, and further improvements will come from using new materials. Perovskite panels are the next up coming down the research pipeline, so it's very cool to hear those are moving out of the research phase and actually hitting production. I expect perovskite-based panels will have years of efficiency gains, just like silicon did, as we continue to explore even more efficient new materials. Great stuff!
[1] Mostly from listening to the Skeptic's Guide podcast, which frequently covers green energy research.
As someone who worked on advanced solar in the 2008-2014 timeframe and later did a PhD in photonic materials, I have learned to heed the old adage "Never bet against silicon".
The cost of silicon solar modules is projected to drop below 10cents/Watt this year and will keep dropping. The major cost of solar is now the installation and grid interconnect. Since these cells have serious problems with lifetime (years at best vs decades for silicon), all indications are that they will be much more expensive at the system level.
In my opinion the major barrier to solar adoption is not efficiency but integrated operation. For instance, my roof has enough area to support a 20kW system, but the utility will only let me put up a 4kW array due because they can't accept the extra energy and stay profitable. This business model problem is not related to efficiency but the result of resistance to distributed energy strategies from utilities who can't understand how to avoid bankruptcy and move away from a centralized power plant based grid.
> the utility will only let me put up a 4kW array due because they can't accept the extra energy and stay profitable.
Transmission congestion might be a more important issue than profitability:
"Avoiding the congestion is essential for a competitive electricity market and is one of the toughest problems of its design." [1]
The course of Damien Ernst [2] gives an excellent overview of all the challenges related to decentralized electricity markets.
Yes this is major component of what I mean by profitability. The utilities spend vast sums on grid equipment and transmission lines that they pay for over decades through bond offerings. The grid is designed to distribute electricity from centralized generation stations to distributed customers (a "mainframe" like model). They support the capital cost of this infrastructure (debt payments) and maintenance on the equipment through their per kWH rates.
An example of this is that, where I live, some depreciated hydro assets produce power at $0.0025/kWh but the electricity rate is $0.11-0.14/kWh. It is not unusual for the majority of the cost of electricity be in debt and equipment maintenance rather than generation.
If I generate electricity on my roof then the utility is screwed from both ends, they must credit me way more than it costs them to generate their own electricity, and feeding electricity into a grid not designed for it adds further wear and tear to the components. Their revenue goes down and their costs go up.
Unsurprisingly, given their sunk costs and the prospect of defaulting on huge bond obligations, they will not permit me to install a rooftop array that will generate more than 40% of my usage, even if paired with large battery systems.
This. I worked for what in our deregulated market is called a "distribution company" (owning the transmission from the HV lines to the meter at your home), and can absolutely attest to what enslavedrobot has said. The funding model for getting electricity from A to B was predicated on assumptions that you could amortize the high capital costs over multi-decades. Which is all fine until distributed generation (rooftop solar) exploded in a time-frame much smaller than expected. Tbh my personal view is that the writing was on the wall large enough for anyone with eyes willing to see to start planning accordingly, but true to form, most companies just tried resisting the change.
I thought infeed limits were all about protecting the grid from power surges; what can an enlightened utility do differently? Battery storage would double the cost of community-generated solar, so even an enlightened utility might be unprofitable taking non-dispatchable power from homeowners, no?
Prices per panel may be dropping but that’s didn’t stop a major panel installer to quote me over $100k for 18kW system with two Tesla batteries. (The installer was not Tesla).
How can one take advantage of cheap panels and have quality work done on the roof?
A lot of it has to do with permitting, building codes, and lack of competition. In Australia deregulation has led to pricing well below 1AU$/W installed. If we could adapt what worked there prices would come down like crazy.
In Australia an 18kW system.installed with two 10kW/h Tesla batteries should run under $30,000 AUD installed. That's somewhere in the vicinity of $23,000 USD.
I've been looking at upgrading my rooftop solar in Australia, since we have had 5kW on the roof for close to 15 years now. Ill put away the pennies for a few more years and pull the trigger at a similar time to when we get our first EV / PHEV with V2G.
Two of those is $27,200 minus installation and some solar panels to feed them (but the PW3 does have 3x 6.6kW solar MPPTs built-in unlike previous ones, so all you need to do is connect the panels directly to the PW3 and then the PW3 to your home).
> How can one take advantage of cheap panels and have quality work done on the roof?
Teach your kids how the blue collar trades are a better deal than a college degree, wait a decade or two, and contractor labor prices should be reasonable again.
The 4Kw limit should be on the infeed and not the solar panel capacity.
You could put 10Kw worth of panels up just limit the output to 4Kw. Now you have a more stable 4Kw feed.
Yeah, my dad has that setup. Over provisioned panels so he still can get max output more of the year. Not to the amount you suggest, but his inverter can safely have about 20% more input than output.
He bought used panels so the actual input may be a little lower than rated (though it doesn’t seem much lower), but he says he sees some ads for new panels nearly as cheap as he paid for used ones 5 years ago.
Is it or is it becoming profitable to have more power on the roof, but dedicated to local use? What local use? Local battery? Bitcoin mining? Aluminum plant (jk)?
With solid state batteries finally hitting production, I'm betting we'll see just that. The more common materials alone will make huge strides once scale-up pains are over
Do you have suggestions for the utilities on how to move away from a centralized grid, avoiding bankruptcy and also providing the same level of reliability as the last few decades?
The only supply side outage that comes to my mind is the Texas cold snap messing with the gas plants.
Perhaps neighborhood level energy storage and an increase in energy transmission costs.
It's very unlikely that a centralized grid will go away, society wants 100% energy availability 100% of the time. So like everything else, people are just going to have to pay for it. The same way you pay for schools even if you don't have kids or pay for roads even if you don't have a car.
It's a tough problem. Trillions of dollars of utilities bonds (mostly owned by pension funds) and a century of regulatory barriers make change hard. The problem is similar to the task of reforming the medical system.
The strategy that I like is to build out distributed systems in "non-integrated areas". These are locations that are not served by the grid and often have a tiny local grid that provides high cost electricity (~1$/kWh). These areas represent test beds for distributed energy tech and might be a place where de-costing and scaling strategies could be developed.
Another strategy is to wait for baby boomers to die. :)
With 4-5 MW batteries in the shape of shipping containers are already now available and rated for 6-10k cycles before much degradation. I think local neighborhood storage near consumption should become common in the near future. Of course electricity distribution companies etc will drag their feet specially in the US with their captured markets and near monopolies.
To me it seems hard to justify use of perovskite panels (outside of niche use cases where high-efficiency is important like space exploration) when they use so many toxic materials (and silicon panels have no such problems and are dirt cheap). Just building more silicon panels seems like a better plan.
Yeah, that's 100% a problem worth calling out and fixing. But we really do need efficiency gains in solar panels, and silicon is maxed out. Figuring out how to make perovskite more workable is absolutely worth the effort, and commercializing it is a crucial step for that.
Do we need efficiency gains? Like more is better, but in the US, land is cheap in many areas. A panel that is 5% better, but degrades faster might not be an economic win for a commercial power plant.
> when they use so many toxic materials ... but degrades faster
At some point, we need to consider that "labour cost becomes dominant" is absolutely irrelevant if the external costs we're completely ignoring at enormous.
> Do we need efficiency gains? Like more is better, but in the US, land is cheap in many areas.
Yes. Anything that lets us offline coal plants faster will save lives. Assuming all else is equal, it's literally free energy. Why wouldn't you want that? What a bizarre question.
> A panel that is 5% better, but degrades faster might not be an economic win for a commercial power plant.
Obviously you have to multiply efficiency by longevity. If that equation didn't work out, they wouldn't be commercializing it.
> Yes. Anything that lets us offline coal plants faster will save lives. Assuming all else is equal, it's literally free energy.
I agree with that. But all else is not equal. The more efficient panels are much more expensive than the regular panels (and there are no signs of this changing any time soon). Cheaper regular panels (and prices are still falling) are more helpful. Aside from which, panels are already cheap enough for cheap energy. It's energy storage costs (and availability of storage technology in general) that are the blocker at this point.
> Obviously you have to multiply efficiency by longevity. If that equation didn't work out, they wouldn't be commercializing it.
I suspect they won't be commercialised for grid-scale power plants. They'll likely be used in space, and perhaps on things like boats and RVs where space is also at a premium.
> The more efficient panels are much more expensive than the regular panels (and there are no signs of this changing any time soon).
I'm pretty confident we'll be replacing silicon panels with another, more efficient tech before long. Something will bust through silicon's efficiency barrier. It may or may not be perovskite-based panels, who knows, buit I still think it's exciting to see the research & gains in this area. It's great to have more options coming online.
It is as free as nuclear, or water generated. The infrastructure must be installed and maintained. Panels, their cleaning, changing failed/broken, inverters, cables, batteries (eventually).
Right now it is mandatory to install in Germany after a major roof renovation. Turns out the typical small home electric needs are about 1000 EUR per year, the installation of a solar system is about 25000. I do not see what is free…
Yes. Should is the key word. As the government pushes lots of people to install solar, prices soar… also my house is particularly bad for solar (roof parts looking exactly west-east) so I install double of what will be used. Also high roof, so according to regulations, all house has to be with scaffolds around. Just that+permits+ connection to the grid by a “meister” costs around 5k… German efficiency is called…
Well, the article under discussion is about the tech being commercialized, so presumably they believe it is cost competitive at least in some scenarios.
I don't understand why you're so hung up on the idea that 120 units for the same cost as 100 units means you get 20 units for free when comparing the two options.
Here, I'll spell it out a bit more:
Assume the mfr is not lying and the panels are 20% more efficient.
Assume that in order to commercialize a new product, it must be cost competitive with existing options. Otherwise no one would choose it, and it would not be commercialize-able.
Therefore, it is a reasonable position to assume that the new panels will give 20% more energy for about the same cost.
This is all hand-wavey, and it is of course possible the commercialization will fail. But until that happens, I think it's pretty cool that we have new tech coming to the market that's showing significant efficiency improvements!
I have no problem with the math, I take issue with your assumptions to get there, namely that these panels will be anywhere near cost-competitive with traditional panels. I hinted earlier that this new panel is manufactured by taking a traditional one and slapping a perovskite cell onto it, so you are assuming this whole tech is literally free.
I think this is amazing tech too, but you're maintaining "this is free energy" with zero evidence outside of a press release that does not mention cost. I'm sorry, this isn't hand-wavey, it's flat-out misinformation. If you have actual information on pricing, please share it.
> so you are assuming this whole tech is literally free.
No I'm not. I'm assuming it's commercially viable, or else they wouldn't be trying to put it into production.
The context of the post you're being weird about was a reply to someone saying "Do we need [solar panel] efficiency gains?", I wasn't specifically talking about the numbers of this tech in that post.
I assumed they were talking about the numbers of this tech in that post. I assumed everyone was talking about the numbers of this tech. You quoted the next bit about land use in the US:
> > Do we need efficiency gains? Like more is better, but in the US, land is cheap in many areas.
> Yes. Anything that lets us offline coal plants faster will save lives. Assuming all else is equal, it's literally free energy. Why wouldn't you want that? What a bizarre question.
I'm sorry if I'm being weird. It really looks like you're arguing efficiency is something standing in the way of saving lives.
> Assuming all else is equal, it's literally free energy.
When comparing silicon and perovskite (and probably any other material) this is a bad assumption. Since this is a bad assumption, the rest of your position falls apart.
There are space limited use cases which would benefit from higher efficiency such as cars and yachts. Yachts are nearly able to cruise entirely on solar.
This is the first I’ve heard of perovskite so I have no idea: are the cited toxicity issues fixable? Or are they inherent to the chemistry of this class of panels? How do we decide how much extra short term toxicity is acceptable? Do we do that blindly without theoretical/empirical results that demonstrate a better end state?
I'm no expert here, either. Lead is regulated pretty carefully these days, so I have to think if this was a deal-killer, then it would've killed the deal already. This conversation has the feel to me of internet armchair commenters insisting they know more than the experts. Shrug.
My understanding is that there could in theory be perovskites that don't contain the toxic materials. But all known perovskites that are suitable for PV panels do happen to contain them.
Because we're already past midnight on the climate catastrophe clock, and every gain counts. If we get a 20% jump in solar panel efficiency, that means we can offline coal plants even more quickly.
Producing more cheap panels is much better solution than waiting for more expensive, more efficient panels that aren't available now. I'm surprised someone who is concerned about catastrophe wants to wait. Solar panels are cheap enough that big installations are putting them on the ground than making mounts. There is plenty of ground.
There are places where more efficient panels would be useful. With rooves, where if spending effort to mount them then might be worth using more efficient panels. Or the flexible panels might be easier to mount.
> Producing more cheap panels is much better solution than waiting for more expensive, more efficient panels that aren't available now. I'm surprised someone who is concerned about catastrophe wants to wait.
Huh? Can you point out where I said we should wait?
It was rhetorical cause you obviously don't. But you are arguing for the slower, more expensive option. Current panels are faster and cheaper. The efficiency doesn't really matter for utility solar.
It is unlikely that perovskite panels will ever get cheap enough for efficiency gain to matter unless there is some breakthrough in production. Current solar panels have too much of a head start.
> But you are arguing for the slower, more expensive option
No I'm not. I'm not arguing for anything. I'm excited to have more options for solar deployments! If the best option for a current time & deployment is silicon, then great! Use silicon! If perovskite does nothing more than put price pressure on silicon to get even cheaper, then that's great, too! I'm excited about green technologies :)
> It is unlikely that perovskite panels will ever get cheap enough for efficiency gain to matter unless there is some breakthrough in production
We'll see! We're pretty good at making improvements. Will perovskite see the same amazing price drop that silicon has? I don't know, but I'm excited to see the first step in that possible path happening.
Solar panels will continues to be installed at a breakneck pace for likely another decade or more. So the sooner we get more efficient panels, the sooner they can be installed rather then the less efficient ones. It's very straight forward what they were saying.
> It's very straight forward what they were saying.
This thread has been a wild exercise in people tripping all over their own feet in their huge rush to educate me that a new technology is not some flawless miracle device, lol.
Armchair comment. Roofs are about the worst place. All the working at heights safety issues, the wide diversity of roof plans, the small size of individual roofs, the cost and difficulty of maintenance, and on and on.
The real move here is to roll them out in fields mostly unusable for other purposes. These days, you can put up an entire field installation in a day or two. Getting them onto roofs requires a whole lot more effort for a whole lot less capacity.
Furthermore you can take a shortcut by mounting them either flat on the ground (dead simple) or vertically (no worries about dust or snow) (use them as a fence). The loss of efficiency might be marginal, depending on your latitude and weather.
Am I wrong in assuming that a 20% efficiency increase in a 25% efficient cell means 5% more energy gained (25*1.2), so a total of 30% efficiency, in relation to the 100% of solar energy which reach the cells?
How much does this efficiency affect levelised battery-backed electricity cost? And how much of the final price is installation, storage, land cost, maintenance and all these other overheads.
My impression is that when the sun shineth, raw solar is almost too cheap to meter... But everything else costs money.
[1] Mostly from listening to the Skeptic's Guide podcast, which frequently covers green energy research.