This would seem to be more common from a business perspective, but apparently it isn’t. This warehouse park, Morgongåva, near Uppsala in Sweden has at least 2.5 MW installed and installing more. This one of their buildings.
It's interesting that they are selling the excess to the local supplier as they will be able to offset the cost faster, and start making a profit from it sooner. That is, the point at which the amount of money saved from not paying for the electricity and the amount gained from selling the excess balances the cost for the installation.
This happens in residential solar a lot. During the Spring/Summer/Autumn you will produce 1.5x - 2x what you consume on any given day but the peak of the power all comes at 9-10am to 2-3pm. You need large amounts of battery to move that power to cover the house for the other 2/3 of the day and that adds a lot of cost and is often similar price to the solar array.
One of the best ways to deal with the problem is selling it back to the grid but how that is done vastly changes the payback period. Some states do it as a unit exchange so you sell 1KWh to them then you get that back later when you want it for no cost. Whereas many do a sale price with a price half or worse. Which price you can get for that power and whether its following wholesale (which tends to be peak early evening and be more during the day than night) can drastically change the pay off period.
So, power when you're not at home, the house is empty, it's warm, and your electric car is with you at work, and no power at night, when you need to cook, heat and charge your car.
This is one of the reasons I've argued in favor of "slow, at work charging" as the default for EVs. If you put in a bunch of 240V/16A chargers (3.8kW), as people show up in the 8-9 timeframe, they start charging, and the bulk of the charging happens in the ~10AM-2PM window that's after the morning peak, as solar is ramping up, and before the afternoon cooling/evening peak. You can even restrict the chargers a bit more in the morning if needed until off peak, though I'm not sure this is worth much over just using dumb chargers that are cheap to install. An average 35 miles a day driving (in the US) requires ~10kWh, so you've got a few hours of charging that can swallow an awful lot of power during the time nothing else is using much of it.
It's far better than having to deal with evening and overnight charging.
But for that you need every parking space in every office building equipped by a charger, and somehow have the grid strong enough to power all that from solar cells which need to be installed somewhere further away, since an average office building does not have enough roof space to charge all those cars. The hours you mentioned are usually spent at work and the cars are parked there (and not at home).
Who said every parking space? I didn't. Even at the coastal FAANG offices, EVs are still the minority. And not everyone will be charging every day. It's still fine to have a lot of non-charging spots - some people have short commutes and can be perfectly fine charging for 8-9h once a week (at 3.8kW, that's around 100 miles in 8h of charging). My experience from the early to mid teens is that the Leafs fought for charging, and the longer range BEVs charged if they could, and otherwise didn't get in the (damned near literal) fistfights over charging, but if someone showed up early, they'd plug in for the day.
If you look at grid demand curves, that "late morning, early afternoon" time spot is the mid-day low for most grids. So there's excess capacity anyway, and if you look at the "duck curve" sort of graphs, there's a ton of solar on the grid then anyway. This is only increasing with time. So we may as well make decent use of it.
You could also have some lower power, 1440W charger slots (15A @ 120V) for those who don't need much power - it'll still make up an average day's driving in 8h of charging, but since it's literally the same wiring cost and such to run 120V as 240V, I'm not sure you gain much with it, and it's now perfectly valid for chargers to coordinate regarding total circuit demand anyway.
> Who said every parking space? I didn't. Even at the coastal FAANG offices, EVs are still the minority.
Might just be me, but I read any comment talking about an ideal future scenario for charging vehicles (if not otherwise qualified) as being for the ideal world where the 99% switched to electric vehicles.
If you're talking about planning for the situation that we have today... that doesn't seem too useful and would be outdated in two-odd years.
It's not an "ideal future".... considering how many countries promised that they'll ban gasoline powered cars by 2035 it's something we should already start building (both the solar side and the chargers).
Compared to all the problems humanity faces (such as not addressing the low-carbon energy transition) it’s hard to be intimidated by a situation where EVs become wildly popular and we have to spend a modest fraction of their cost on wiring charging infrastructure and grid upgrades (a cost we can amortize over decades.)
It's not just the cost... it's the material and time needed, all the regulation, permits, etc.
Forbidding something is easy... but where the hell will people charge their cars before the cargers are installed? Do you know how long it takes to build a power plant? Do you know how much those things cost? Is there even a company able to build literally billions of charging stations in the next 20 years? 2035 is comming fast... some people have been waiting for a basic build permit for longer than that.
I did... if you want to stop selling gas powered cars by 2035 (or whatever the current cutoff date is), you need pretty much every parking space fitted with a charger in ~2040, since electric should overtake gas powered by then.
High power, low power,... it doesn't matter if you don't have physical chargers installed pretty much everywhere, and enough solar installed to cover the power usage. This is not "distant future", this is less than 20 years away. Suburbia is simple, because most people have an outlet in their garage, but they can only charge at night (no solar), apartment buildings have basically the same issues as office car parks, but with energy needed at night, and if you really want to utilize solar, you need chargers at office car parks too.
Your house can act as a thermal battery and draw power to heat or cool it while there is a surplus. You can also have a hot water reservoir as another large thermal battery for showering etc.
Where I am, Sweden, last year I got 3x the price during the day for my sold electricity, than what I paid at night. So car charging at night and selling electricity during the day.
Has anyone done more detailed numbers on costing of roof mounted installation vs just putting utility scale solar on the ground?
I would have thought if the economics were super favourable this would already be happening by now.
To me it -seems- like a great idea but while I'm pretty versed in solar for home scale (diy off-grid setups etc) I have no idea how the numbers bear out for this install size, grid connection, etc.
> Has anyone done more detailed numbers on costing of roof mounted installation vs just putting utility scale solar on the ground?
You're far cheaper on the ground, for a variety of reasons (at least in most areas). A good ground mount install should be able to come in around $1/W, and you'll have a hard time getting roof mount below closer to $1.50 or $2/W installed, and if you get $2/W, you're getting a great deal.
If you're doing the work yourself, ground mount for around $1/W is doable (including the frames), but roof mount I've not been able to get below about $1.25/W, and that's hard - $1.50/W is more typical.
The main difference is that the roof mount system requires rapid shutdown on (in NEC 2017 and later) every panel, and you're limited to 600V (though with rapid shutdown requirements, this is less relevant). For a ground mount system that's isolated away from random people (fenced area), you can run up to 1500VDC to the inverters, and this rather reduces your costs in wiring. You also don't need rapid shutdown, so you can just run strings of panels (20 or 25x 72 cell panels in series, depending on the environment).
Also, if those strings have an arc fault somewhere, you're not going to burn up much that matters for a big ground mount install. You'll cook a couple panels, and that's about it. A fire under the panels on a warehouse is a much bigger deal.
The main problem you run into with warehouses, though, is that there's just no good option for mounting. For a flat roof, they're generally built to exactly the loading requirements for the area. You don't have the spare PSF capacity for a ballasted mount, which means that assuming you've got the weight rating for the panels, you're looking at a lot of roof penetrations on a flat membrane roof to hold stuff down - and the odds of some of those leaking is basically 100%.
Plus it's a royal pain to work on roof mount systems.
Thanks for the great writeup! The root loading makes perfect sense when you think about it and in that context makes the equation much more expensive unless the warehouse could feasibly use all the power generated (which seems unlikely).
Do you know how the numbers compare for green field projects on the ground? They keep building new projects in otherwise wild/wilderness areas instead of on rooftops.
we massively under-value land in most cases, which is the reason that activists push for solutions like this.
utility scale solar on the ground is economically better, if you're looking at "the numbers". but utility scale solar probably means repurposing farmland, because that's the cheapest land that has enough infrastructure built up around it to make a utility install easy. and repurposing farmland to build solar isn't a clear win in terms of overall societal benefit.
> ... but utility scale solar probably means repurposing farmland...
Great. Let's start by repurposing the land that grows the corn we turn into ethanol at somewhere around or below unity EROEI, as a political handout to the midwestern states. "Burning a megajoule of diesel, to grow enough corn to make something slightly less than a megajoule of gasoline," is not a winning solution to any problem but "How to best hand money to states that happen to have early Presidential primaries." It's not a bad gig for Iowa (I lived there for a decade), but neither is it a particularly good use of energy, fossil fuels, or money. Let's put solar up there, and work out.
Once we've gotten rid of the ethanol debacle, we can see where things are and discuss from that point, but there is a lot of farmland right now that doesn't grow human or animal food, it grows corn that we process rather inefficiently into ethanol.
i completely agree, but as long as this is all driven by economics the existence of those corn subsidies is going to mean the farmland that gets repurposed won't be the farmland that gets the most government subsidies, it'll be the farmland that grows the cheap food that people need to eat.
i know ethanol farming is a waste, but it's a lot easier to convert a corn field back into a useful farm field before you've built a solar farm on top of it, rather than afterwards
Converting farm land into solar, or even fallow land, seems like a massive societal benefit. Most farmland is fantastically unproductive, used for excess corn, soy, or sugar that we struggle to find uses for.
Taking even a fraction of the land used for ethanol and using it for solar would give us enough energy to power the US. And the ethanol business is a complete make work program.
ya till solar hits a tipping point like it is in California right now. Then the utilities want to set a flat rate based on wages to stay connected to the grid.
This will 100% work but we also need battery tech to hit like 250-300 a kwh for storage is my magic number to get a 40-50kwh backup and detach from the grid.
The storage market has been revolutionized in the past few years thanks to LFP chemistries, and it's only going to get better soon. There are massive new US production facilities coming online, and Bloomberg is estimating that US cell costs will be $83/kWh, just a few dollars more than Chinese costs. Add in the IRA tax credits, and the cost will be less than $50/kWh. Packs will be more expensive than cells, of course, but the future looks very rosy. Could see that pack at half its current price, even in quantities as small as 50kWh.
I always chuckle when I encounter the sweet summer children that think that battery tech is stagnant...
It's an extremely common argument from renewable opponents that storage is just too expensive and will never be economical. If you manage to get them to write a number down it's always something like $500-$1000/kWh because that's what some old fossil fuel study from the 90s used.
It's likely far cheaper than buying the electricity from the grid, but it will depend on local commercial electricity rates. Utilities usually have to set up pretty perverse schemes in order for solar to not be an economical choice.
The short answer is "No, but it sounds obvious if you don't know much about it." Similar to Solar Roadways - the less you know about solar or roads, the better the idea sounds. The half mile gouge down a road near me from a trailer coming free and riding on the chains until they noticed and pulled over isn't a big deal on asphalt (they'll fill it next time they chipseal it), but it would be millions of dollars worth of damage to solar road tiles. Among many other issues.
It's the sort of thing that sounds obvious if you don't know much about the field, and the more you learn, the more things you realize make it a royal pain in the rear to deal with unless the structure was designed with solar in mind - and even then, it's a pain compared to a ground mount array somewhere close.
For a big membrane roof, you don't want a ton of penetrations - which means ballasted mounts. Except they weigh a lot, and the roof usually isn't built to support them on top of the required weight for rain/snow/etc. So you need to screw the stuff down to the roof, but now you have tons of penetrations in a membrane roof, and if anyone tells you they can do thousands of those without a single leak, they're full of crap.
As I talked about in another comment, you can't just run high voltage DC strings on a roof - you need per panel electronics, so the normal solution is microinverters, but now you're pushing 240V around instead of the 1500V you can run on ground mounted strings, so wiring cost is higher, and Enphase sure makes their money off their handout in the NEC 2017. Reliability of microinverters remains an open question as well. They're not exactly in a good spot for electronics.
You've got open area, you've got a high current interconnection, but the rest of it is just a set of thorny problems that makes it quite a bit more expensive than ground mount.
I've no idea why you're being downvoted, because it is a very real issue. You don't have to burn down many warehouses with solar before insurance companies start to have a lot of hard questions and rate increases. In tens of thousands of connections, it doesn't take a high failure rate on components before you've got something high resistance.
Could you elaborate? I feel like I’m missing something significant. I am imagining people file out, then file back in a few hours later, which seems harmless-ish.
I would assume downtime in terms of "we're paying all these warehouse workers to stand outside - now we have to send them home and the warehouse is unusable, we don't know if the product inside is salvageable due to potential water damage, etc. etc."
Probably a lot easier on the business risk side of things to not have solar on the roof because then the risk for a fire is massively reduced in the first place.
Exactly this. Depending on the size of the operation the financial impact of a single day of delay might offset savings from many years. Specially when you could've put solar somewhere else other than on top of your huge warehouse, if you own a huge warehouse.
When we installed solar, Tesla's was the most expensive out of all the proposals. Their solar roof looks very very cool, but it's definitely not "cheap stuff".
Never heard of anything like that, but most of my experience is from 2+ years ago. Is your install recent? Are you sure they were equivalent proposals? If you add storage, that's no longer just solar, and their storage offerings are far more expensive than others.
I couldn't find a local solar installer that wanted to use the type of cheap inverters that Tesla wanted to use. My neighbor has to have Tesla come out about once a year to fix issues.
The first is the "solar roof" sort of thing that... we'll see how maintainable it is, but you get to write a properly large check for it, and most of the people I know with it have a couple Powerwalls, and are quite adverse to sharing any sort of actual pricing on it, muttering about "Well, wife acceptance factor" and "But it'll be worth it in power outages..." and such. In other words, almost certainly north of $5/W, and they likely didn't get an itemized invoice of materials/labor/etc.
Tesla also, in some markets, has a "standard solar panel" offering that typically comes in around $2.50/W, give or take, and... well, you get what you pay for, typically. It ranges from "decent and a good deal" to "a hot mess," and you've no idea what you're going to get ahead of time.
... and then there are people like me who have sub-$1.50/W installs (DIY). Mine came in around there, but after figuring out how to optimize it better, a neighbor's build came in around $1/W for a 21kW system.
//EDIT: And, yes, it takes a lot of learning to do. Cross that with the tens of thousands of dollars saved, I think it's well worth the hassle.
$2,50/Wp seems very high! Dutch market is somewhere between 0.8-1.5€/Wp installed, and we’re not exactly a low wage country. (Small commercial can be as low as 0.65€/Wp, at around 250 panels)
Any ideas on what’s driving the cost for installs in the US compared to the EU? I just drove through Florida on vacation and it’s mind boggling to me how few roofs have PV here.
Florida doesn't have overwhelming incentives for solar roofs, although I believe our governor vetoed a change to the rules last year. Also, many people will replace their roof at the same time that they get solar which drives up the cost. There's also the issue of roof maintenance in Florida because our weather is super extreme and necessitates the replacement of asphalt shingle roofs around every 15-20 years or so due to insurance costs.
When you did your DIY install, was that for everything (purchasing the necessary parts, wiring, roof mounting, etc), or did you DIY certain parts of it and leave the rest up to professionals?
"Late last year, while CEA was still conducting its inspections, it informed Amazon of one critical and 259 major findings across Amazon's rooftop solar portfolio. Problems included mismatched module-to-module connectors, improper installation of connectors, poor wire management and evidence of water intrusion in the inverters, internal documents said. "
That's not a solar problem, that's an installer problem. Clearly you need to have well-qualified people doing and regularly inspecting the installing, just as you would with any type of large-scale infrastructure. Bridges also collapse when they're not inspected, airplanes fall out of the sky.
The quality of solar tech varies immensely and it will likely take regulation to address some of these issues. But I do not think this is the primary concern here.
We put solar on our house and did some pretty rigorous research into the tech precisely because of this concern. After spending many nights/hours understanding the tech in this industry, I can confidently say it matters less about the current generation of solar tech. But, while the degree of quality in solar products exist, what really matters are the contractors that install it. I did not read the articles and I don't want to claim that there's direct causation by contractors, but if I were a betting man, I'd easily put money on the quality of the installation as opposed to the actual tech provided.
Structural design issues are the problem as other comments note. If you haven't designed the roof from the beginning to have a fairly weighty additional structure attached on top, then all kinds of other problems can arise, from leaks to wind-related loads and so on.
Over the long term, it becomes something building codes need to implement, similar to electrical wiring and plumbing standards:
NJ has code that requires structural loading on new warehouses to be able to support solar on 40% of the roof. It’s just a matter of getting this into code as soon as possible, so new buildings are solar ready as old warehouse space is eventually retired at EOL.
Yes, and in areas that get snow the roof would need to be strong enough to support the max snow weight plus the weight of the solar gear. I don’t know how much safety factor most buildings have, but many might require additional reinforcement.
I know of a city building here where they wanted to install solar and as they got into it realized they would need to hire a steel company to come in and reinforce all the steel roof trusses. That was custom work with welders up on scaffolds and a lot of other work to remove lighting and other fixtures to give them access (and then reinstall it all afterwards). They went ahead with it but with the extra expense it will never be paid back in electrical savings.
By the way, this was the thesis of the unfairly* maligned Solyndra: their collectors were extremely light and encased in tubes. The idea was that they were light and could be anchored without through holes to joists (every through hole in a roof is a leak opportunity for a leak, especially when it’s a retrofit; with the weight of panels on a flat industrial roof you often have to anchor them to a structural member). Also the tube structure permitted airflow so there wasn’t a heat build up nor risk of a “wing” phenomenon.
* one political party chose to use Solyndra as a cudgel against a reasonable industrial program. I was always dubious that their thin film approach would pan out, and indeed it didn’t which was what killed the company. But admired their basic thesis, which I still think was sound.
For a while, Batteryhookup was selling pallets of Solyndra modules for unbelievably cheap. I think it's difficult to use them on buildings now because the NEC's new rapid-shutdown requirements effectively made decades of solar stuff obsolete (welp so much for the environment), and they're not useful for making solar carports and stuff. But maybe someone with a lot of land...?
Anyway, it's a shame. They were a really neat concept and sidestepped a lot of issues that raise the cost of other systems.
This misses the point. The limiting factor for solar isn't real-estate. It is inverter production, battery capacity, interconnect, ... basically all of the parts of turning solar energy into grid electricity other than the land for solar panels.
The core obstacle is financing - solar traditionally takes a long time to pay back from energy savings, so the financials of solar on warehouses didn't work out. The IRA made it possible to sell the tax credits from solar deployment on warehouse roofs (e.g. https://www.reunioninfra.com/) so now solar on warehouse rooftops can make financial sense in many more instances.
How are disputes about roof maintenance settled among a roof tenant (the energy company), the tenant of the warehouse, and the owner of the building? If a roof is leaking but repairing the leak will interfere with energy production, how are repairs prioritized? This is an easier situation to handle when there is no energy tenant nor warehouse tenant but rather an owner-occupied who is also the energy provider.
It all depends how the leases are written and who's responsible for the solar panels. Since the panels could potentially be the landlord's, warehouse tenant's, or a third party roof tenant's, it could be written up in a variety of ways. The Landloard is going to be the most concerned about it as A) it could potentially effect the base building, and/or B) tenants may take issue with a leak. Chances are the repair would be rolled up into CAM (Common Area Maintenace) charges, which are generally dispersed to the tenant based on an arrangement in the lease (typically % of cost based on % of sq ft of the property).
The best chance for it to be handled some other way is if the warehouse tenant and the solar tenant are one in the same. The landlord will still likely be involved, but chances are the tenant will do the repair.
When you see something like [1] in the center of a city of 10 million people, that's a lot of potential for energy generation. And the great thing about putting a ton of panels in the middle of the city is it tends to avoid all the transmission issues we have with siting PV facilities way out in the desert.
67 comments
[ 3.9 ms ] story [ 130 ms ] threadhttps://renewablesnow.com/news/swedish-online-pharmacy-apote...
One of the best ways to deal with the problem is selling it back to the grid but how that is done vastly changes the payback period. Some states do it as a unit exchange so you sell 1KWh to them then you get that back later when you want it for no cost. Whereas many do a sale price with a price half or worse. Which price you can get for that power and whether its following wholesale (which tends to be peak early evening and be more during the day than night) can drastically change the pay off period.
This is one of the reasons I've argued in favor of "slow, at work charging" as the default for EVs. If you put in a bunch of 240V/16A chargers (3.8kW), as people show up in the 8-9 timeframe, they start charging, and the bulk of the charging happens in the ~10AM-2PM window that's after the morning peak, as solar is ramping up, and before the afternoon cooling/evening peak. You can even restrict the chargers a bit more in the morning if needed until off peak, though I'm not sure this is worth much over just using dumb chargers that are cheap to install. An average 35 miles a day driving (in the US) requires ~10kWh, so you've got a few hours of charging that can swallow an awful lot of power during the time nothing else is using much of it.
It's far better than having to deal with evening and overnight charging.
If you look at grid demand curves, that "late morning, early afternoon" time spot is the mid-day low for most grids. So there's excess capacity anyway, and if you look at the "duck curve" sort of graphs, there's a ton of solar on the grid then anyway. This is only increasing with time. So we may as well make decent use of it.
You could also have some lower power, 1440W charger slots (15A @ 120V) for those who don't need much power - it'll still make up an average day's driving in 8h of charging, but since it's literally the same wiring cost and such to run 120V as 240V, I'm not sure you gain much with it, and it's now perfectly valid for chargers to coordinate regarding total circuit demand anyway.
Might just be me, but I read any comment talking about an ideal future scenario for charging vehicles (if not otherwise qualified) as being for the ideal world where the 99% switched to electric vehicles.
If you're talking about planning for the situation that we have today... that doesn't seem too useful and would be outdated in two-odd years.
Forbidding something is easy... but where the hell will people charge their cars before the cargers are installed? Do you know how long it takes to build a power plant? Do you know how much those things cost? Is there even a company able to build literally billions of charging stations in the next 20 years? 2035 is comming fast... some people have been waiting for a basic build permit for longer than that.
I did... if you want to stop selling gas powered cars by 2035 (or whatever the current cutoff date is), you need pretty much every parking space fitted with a charger in ~2040, since electric should overtake gas powered by then.
High power, low power,... it doesn't matter if you don't have physical chargers installed pretty much everywhere, and enough solar installed to cover the power usage. This is not "distant future", this is less than 20 years away. Suburbia is simple, because most people have an outlet in their garage, but they can only charge at night (no solar), apartment buildings have basically the same issues as office car parks, but with energy needed at night, and if you really want to utilize solar, you need chargers at office car parks too.
I would have thought if the economics were super favourable this would already be happening by now.
To me it -seems- like a great idea but while I'm pretty versed in solar for home scale (diy off-grid setups etc) I have no idea how the numbers bear out for this install size, grid connection, etc.
This is usually under "C&I" (commercial and industrial), as opposed to rooftop or utility scale.
Try Lazard LCoE report, NREL Solar reports, IRENA renewables, or use google.
You're far cheaper on the ground, for a variety of reasons (at least in most areas). A good ground mount install should be able to come in around $1/W, and you'll have a hard time getting roof mount below closer to $1.50 or $2/W installed, and if you get $2/W, you're getting a great deal.
If you're doing the work yourself, ground mount for around $1/W is doable (including the frames), but roof mount I've not been able to get below about $1.25/W, and that's hard - $1.50/W is more typical.
The main difference is that the roof mount system requires rapid shutdown on (in NEC 2017 and later) every panel, and you're limited to 600V (though with rapid shutdown requirements, this is less relevant). For a ground mount system that's isolated away from random people (fenced area), you can run up to 1500VDC to the inverters, and this rather reduces your costs in wiring. You also don't need rapid shutdown, so you can just run strings of panels (20 or 25x 72 cell panels in series, depending on the environment).
Also, if those strings have an arc fault somewhere, you're not going to burn up much that matters for a big ground mount install. You'll cook a couple panels, and that's about it. A fire under the panels on a warehouse is a much bigger deal.
The main problem you run into with warehouses, though, is that there's just no good option for mounting. For a flat roof, they're generally built to exactly the loading requirements for the area. You don't have the spare PSF capacity for a ballasted mount, which means that assuming you've got the weight rating for the panels, you're looking at a lot of roof penetrations on a flat membrane roof to hold stuff down - and the odds of some of those leaking is basically 100%.
Plus it's a royal pain to work on roof mount systems.
utility scale solar on the ground is economically better, if you're looking at "the numbers". but utility scale solar probably means repurposing farmland, because that's the cheapest land that has enough infrastructure built up around it to make a utility install easy. and repurposing farmland to build solar isn't a clear win in terms of overall societal benefit.
Great. Let's start by repurposing the land that grows the corn we turn into ethanol at somewhere around or below unity EROEI, as a political handout to the midwestern states. "Burning a megajoule of diesel, to grow enough corn to make something slightly less than a megajoule of gasoline," is not a winning solution to any problem but "How to best hand money to states that happen to have early Presidential primaries." It's not a bad gig for Iowa (I lived there for a decade), but neither is it a particularly good use of energy, fossil fuels, or money. Let's put solar up there, and work out.
Once we've gotten rid of the ethanol debacle, we can see where things are and discuss from that point, but there is a lot of farmland right now that doesn't grow human or animal food, it grows corn that we process rather inefficiently into ethanol.
i know ethanol farming is a waste, but it's a lot easier to convert a corn field back into a useful farm field before you've built a solar farm on top of it, rather than afterwards
Or you can do agrovoltaics and have both. There's some evidence that low density agrovoltaic setups increase corn yields.
Taking even a fraction of the land used for ethanol and using it for solar would give us enough energy to power the US. And the ethanol business is a complete make work program.
This will 100% work but we also need battery tech to hit like 250-300 a kwh for storage is my magic number to get a 40-50kwh backup and detach from the grid.
Here you go, a 30kWh 48V battery for $9k ($300/kWh): https://signaturesolar.com/eg4-lifepower4-lithium-batteries-...
I always chuckle when I encounter the sweet summer children that think that battery tech is stagnant...
(Subscription required) https://www.bnef.com/shorts/16235?e=Insight%20Alert:sailthru
It's the sort of thing that sounds obvious if you don't know much about the field, and the more you learn, the more things you realize make it a royal pain in the rear to deal with unless the structure was designed with solar in mind - and even then, it's a pain compared to a ground mount array somewhere close.
For a big membrane roof, you don't want a ton of penetrations - which means ballasted mounts. Except they weigh a lot, and the roof usually isn't built to support them on top of the required weight for rain/snow/etc. So you need to screw the stuff down to the roof, but now you have tons of penetrations in a membrane roof, and if anyone tells you they can do thousands of those without a single leak, they're full of crap.
As I talked about in another comment, you can't just run high voltage DC strings on a roof - you need per panel electronics, so the normal solution is microinverters, but now you're pushing 240V around instead of the 1500V you can run on ground mounted strings, so wiring cost is higher, and Enphase sure makes their money off their handout in the NEC 2017. Reliability of microinverters remains an open question as well. They're not exactly in a good spot for electronics.
You've got open area, you've got a high current interconnection, but the rest of it is just a set of thorny problems that makes it quite a bit more expensive than ground mount.
https://www.cnbc.com/2022/09/01/amazon-took-solar-rooftops-o...
https://www.cnbc.com/2019/11/05/tesla-and-walmart-settle-ove...
Might have been a Tesla issue in both cases.
Probably a lot easier on the business risk side of things to not have solar on the roof because then the risk for a fire is massively reduced in the first place.
I couldn't find a local solar installer that wanted to use the type of cheap inverters that Tesla wanted to use. My neighbor has to have Tesla come out about once a year to fix issues.
The first is the "solar roof" sort of thing that... we'll see how maintainable it is, but you get to write a properly large check for it, and most of the people I know with it have a couple Powerwalls, and are quite adverse to sharing any sort of actual pricing on it, muttering about "Well, wife acceptance factor" and "But it'll be worth it in power outages..." and such. In other words, almost certainly north of $5/W, and they likely didn't get an itemized invoice of materials/labor/etc.
Tesla also, in some markets, has a "standard solar panel" offering that typically comes in around $2.50/W, give or take, and... well, you get what you pay for, typically. It ranges from "decent and a good deal" to "a hot mess," and you've no idea what you're going to get ahead of time.
... and then there are people like me who have sub-$1.50/W installs (DIY). Mine came in around there, but after figuring out how to optimize it better, a neighbor's build came in around $1/W for a 21kW system.
//EDIT: And, yes, it takes a lot of learning to do. Cross that with the tens of thousands of dollars saved, I think it's well worth the hassle.
Any ideas on what’s driving the cost for installs in the US compared to the EU? I just drove through Florida on vacation and it’s mind boggling to me how few roofs have PV here.
I think it’s linked to installation. Some sources say they didn’t remove them after the fires
That's not a solar problem, that's an installer problem. Clearly you need to have well-qualified people doing and regularly inspecting the installing, just as you would with any type of large-scale infrastructure. Bridges also collapse when they're not inspected, airplanes fall out of the sky.
We put solar on our house and did some pretty rigorous research into the tech precisely because of this concern. After spending many nights/hours understanding the tech in this industry, I can confidently say it matters less about the current generation of solar tech. But, while the degree of quality in solar products exist, what really matters are the contractors that install it. I did not read the articles and I don't want to claim that there's direct causation by contractors, but if I were a betting man, I'd easily put money on the quality of the installation as opposed to the actual tech provided.
Over the long term, it becomes something building codes need to implement, similar to electrical wiring and plumbing standards:
https://www.nrel.gov/state-local-tribal/blog/posts/solar-rea...
https://www.jdsupra.com/legalnews/new-jersey-solar-ready-req...
I know of a city building here where they wanted to install solar and as they got into it realized they would need to hire a steel company to come in and reinforce all the steel roof trusses. That was custom work with welders up on scaffolds and a lot of other work to remove lighting and other fixtures to give them access (and then reinstall it all afterwards). They went ahead with it but with the extra expense it will never be paid back in electrical savings.
* one political party chose to use Solyndra as a cudgel against a reasonable industrial program. I was always dubious that their thin film approach would pan out, and indeed it didn’t which was what killed the company. But admired their basic thesis, which I still think was sound.
Anyway, it's a shame. They were a really neat concept and sidestepped a lot of issues that raise the cost of other systems.
The best chance for it to be handled some other way is if the warehouse tenant and the solar tenant are one in the same. The landlord will still likely be involved, but chances are the tenant will do the repair.
1: https://www.google.com/maps/@34.0297809,-117.9823821,3248m/d...