Many houses have solar panels, but in a power outage almost all of them shut down. Especially with serious blackouts like Texas had or like Europe may have this winter, getting even minimal power during an outage is valuable. Designing residential grid-tied rooftop solar so that it only worked when the grid is up was a serious mistake that squandered what could have been a major benefit of putting panels on roofs. Let's fix this!

When the grid goes down it's important that solar panels don't keep sending out power, which could shock utility workers fixing things. Having inverters shut themselves off is the easiest way to ensure that never happens, but it's not the only way: the inverter could disconnect from the grid while providing power just to the house. This would only work when the sun was shining, of course, and only when there was enough sunlight on the panels for what you were trying to run. A large system in a sunny area could be set up to try to power the whole house, while a less productive system might be set up to power a few priority circuits or even just a single outlet.

Making systems work this way by default would have major advantages in a range of disasters. You would immediately have widespread distributed generation, and so many things rely on electricity:

  • Communications. Really important for people to coordinate. Even just a small amount of power for recharging phones would go a long way.

  • Medical devices. Many people rely on ventilators, powered wheelchairs, etc, and these generally only have a few hours of battery if the grid goes down.

  • Refrigeration. Keeping food (and medicine) from spoiling.

  • Heating. Even gas and oil heat need electricity for blowers, pumps, etc. and generally will not function at all when the power is out.

  • Cooling. Fans, AC, etc. Combines especially well with solar, since cooling need is highest when available solar is highest.

  • Pumps, tools, etc. Even more important during disasters.

Looking ahead to likely winter power outages in Europe, and potential outages here if things get worse [1] it would be far better if these millions of rooftop solar installs were able to operate independently.

You can install systems with batteries, which would be more robust than relying on however much power the sun happens to be producing at the moment, but large batteries are expensive and I'm not aware of anyone selling systems like this with small batteries. What if we used no battery?

Electrically, this is pretty practical. When we got solar I thought about this a lot and we installed an SMA Sunny Boy with a "Secure Power Supply" (SPS) in the basement. It didn't end up costing much more: something like $250.

Unfortunately these aren't allowed anymore for residential rooftop solar. Starting with the 2014 National Electrical Code (NEC) you need to have some form of rapid shutdown to protect firefighters in an emergency, and in the 2017 revision it got much stricter to where the SPS functionality would have to be completely redesigned to qualify. The only option now, as far as I can tell, is the recently released Enphase IQ8 Sunlight Backup. It's very expensive: I got some quotes and it added about $7k (23%).

I think it's likely that this rule change didn't consider the resiliency downsides of prohibiting the current versions of these systems, and it would have been better to make the change over a longer period with more notice to manufacturers. For example, it took until December 2021 for the first NEC 2017 compliant backup system, the IQ-8 to come out, 5 years from code publication. One way to get most of the benefits of the code change without this particular harm would have been to offer a temporary exception for island-capable solar.

My understanding is that the basic electronics to support best-effort power are not actually expensive, however, and almost all of what makes it costly is low volume. You could fix the volume problem and scale up the benefits by requiring this functionality as a condition for receiving residential solar subsidies: resiliency is something the government has a strong interest in, and that people are unlikely to fully account for since they think the government has it covered. It's too late to go back and require this from the beginning, but what if we announced that starting in 2026 residential solar would only be subsidy-eligible if it supported best-effort power during blackouts?


[1] I recently bought a generator, just in case. It's a dual fuel one (gasoline or propane) because it's helpful to be able to use multiple options and propane doesn't require gasoline's level of cleanup from intermittent use.

New to LessWrong?

New Comment
32 comments, sorted by Click to highlight new comments since: Today at 9:00 PM

Right on! I think if people had done a cost-benefit analysis in the tradeoffs between risks of worker injuries / deaths versus risks of death / injuries / upheaval during blackouts (and especially extended blackouts), they wouldn’t have set the standards the way they did. But really I bet the decisionmakers were putting 100% of their attention on the first type of risk and 0% of their attention on the second type of risk.

I’m not planning to do anything to help advance this cause, but I very much hope somebody does. (I would sign a petition or whatever.)

How good is the power quality, if you're making best-effort power from solar panels with neither a grid tie nor any associated batteries? If it looks like it's working but provides brownout power frequently, I imagine it could damage downstream electronics enough to not be worth it. OTOH the amount of batteries needed to prevent this would would be quite small compared to the amount of batteries needed to power the house for any meaningful duration.

It's worth noting explicitly: the resiliency advantages are larger than the fraction-of-residences-with-panels might suggest, since in the event of an extended blackout, people with island-capable solar can offer battery-recharging services to their friends and neighbors.

How good is the power quality, if you're making best-effort power from solar panels with neither a grid tie nor any associated batteries? If it looks like it's working but provides brownout power frequently, I imagine it could damage downstream electronics enough to not be worth it.

At least as implemented by SMA, it doesn't brown out. You either get the full voltage or overcurrent protection trips and it needs to cycle. If you draw much less than the solar array is currently able to power you get steady current.

the amount of batteries needed to prevent this would would be quite small compared to the amount of batteries needed to power the house for any meaningful duration.

I agree that something integrating even a very small battery would be great, and I don't know why no one offers this.

Out of curiosity: How is the situation with several parties sharing one solar farm? Since you're sharing the house with several other inhabitants, how do you share the electricity bill? Do you have any form of metering on the different rooms?

We just rent to people "utilities included". Even though the house is two units and two lines of service, it's not wired with everything that supports one unit on one service, so this is what MA requires. While this does mean incentives aren't quite aligned it hasn't been a problem: electricity costs are pretty low compared to rents here: we pay ~$190/month in electricity and collect ~$3.2k/month in rent (source).

If you're, say, roommates in a house that has solar panels, you can do what most people do and split the electricity bill evenly - it's just that, some months, your electricity bill will negative and you'll all get a payout.  If you're in a condo or other situation where you share ownership of the roof and the solar panels with a household with another electrical meter, you'd have to work out sharing the profits/ cost reduction, but you could do it if you wanted to.

That seems like it wouldn't handle cases where capital expenses (buying the system) weren't split evenly?

Good point - I'm not sure how to handle that off hand but people have been involved in business ventures where they have put in different amounts of capital for centuries, people could probably figure it out.

If a solar panel hooked up to your house's electricity provides 1/3 of the power your appliances expect, what happens? Do they not run? Do light bulbs light up more dimly? Or do they break in some way?

With the inverter we have (SMA SunnyBoy with SPS), if you try to draw more power than is available the power goes off. Then you unplug some things, push a button on the inverter, and try again. I expect this is the way all these systems work, since providing less power than things expect is a good way to damage them.

Is it possible to contact the people who write the NEC? This might be the sort of thing where they’d say “oops, didn’t think of that” (or not, idk)

They do have a process where they seek public comments for changes: https://www.atkore.com/industry-affairs/national-electrical-code-process-explained/

You likely could make a new proposal to change the rules. 

Jefft: obviously the Tesla powerwall has offered this for years, it is just expensive.

What you probably haven't heard of is there are now Chinese off brand "powerwalls" that work quite well.

It's a combination of 2 things:

An "all in one" offgrid inverter https://signaturesolar.com/shop-all/off-grid/inverters/ (notice how the prices are not any more than you paid for your sunny boy.  off brand though)

And a "server rack" battery https://signaturesolar.com/shop-all/batteries/eg4-lifepower4/ , which has made battery backup actually affordable at ~$300/kwh.  This is 1/3 to 1/5 the price it was previously, and these are LFP batteries, not lead acid, so they will last 15-25 years.  

Electrically, wiring one of these in is not difficult and some of them are UL listed and legal in most areas. 

The way it works functionally is simple: your 'all in one offgrind inverter' powers a subpanel, where most/all your houses circuits are on.  It receives power from your main panel, and also solar panels, and it is wired to a battery.

During most of the day, everything in your house is directly powered by solar.  Whenever electrical load exceeds solar output and at night, the power comes from the batteries.  A prolonged period of low solar output, you configure the all in one to enable battery charging at ~10% remaining state of charge.  You can also configure it to only charge during the cheapest time of each day.  

You never feed power back to the grid under any circumstance.

You also can wire in a backup generator and relays will cause it to auto-start as needed.  

Somewhere like San Diego a system like this will pay for itself extremely rapidly (<4 years) because of the exotically high electric rates.

obviously the Tesla powerwall has offered this for years, it is just expensive

You may already know this, but making it explicit: grid-tie solar with battery backup is a market that predates the powerwall. Tesla just has the best marketing ;) The main problem, as you said, is that this is expensive.

You never feed power back to the grid under any circumstance.

That's a big downside to this system. The main subsidy for solar has been through generation credits (we're on SREC II; new installs here are SMART) and very favorable net metering rules.

Well in that case, go install an interlock kit and a generator inlet so your generator can be used conveniently.  For emergency situations a generator is just fine.  The fuel cost won't be significant, generators are cheap, even a model big enough to run an entire house with efficient air conditioning.  (9 kilowatt or so dual fuel propane generator)

Get a duel fuel generator so you can store fuel for a decade as propane rather than for 30-90 days in the form of gasoline.  (diesel is too expensive as a generator to bother with)

Your EPS was never a good source of backup power.  Long term everyone is going to have a setup more like I described above with a battery,  you're receiving a large government subsidy in your favor through your net metering deal.

Your EPS was never a good source of backup power.

Why? In a very long outage it seems to me like it does much better than generators which need fuel.

Get a duel fuel generator so you can store fuel for a decade as propane

I do now also have a dual fuel generator, as I say at the end of the post. And I'm storing propane.

rather than for 30-90 days in the form of gasoline

You can store gas for a year or longer if you use fuel stabilizer. I do this as well, primarily as backup for the car.

I do now also have a dual fuel generator, as I say at the end of the post. And I'm storing propane.


    Sorry I missed that part.  Great

You can store gas for a year or longer if you use fuel stabilizer. I do this as well, primarily as backup for the car.

    Technically you can.  Issues: 

  1.  fuel stabilizer makes that claim on the bottle but it's a risky thing to find out it didn't work the one time you actually need an emergency generator.
  2. 10 years (only reason for this limit is this is the rated service life of a propane tank) > 1-2 years
  3. Generators use a mechanical low cost fuel mixing system called a carburetor.  After the engine is run, say to test the generator, gasoline is still in the carb.  It evaporates over time and has a very high chance of causing later failure.  If you are going to use gasoline, run the generator dry when you run it..  (run it until it stops from fuel exhaustion).

1: Looking, I don't see people having problems with stabilized fuel after ~1y -- is this a problem you've seen?

2: I agree it's nicer to be able to just hold onto the propane tank for a decade, but needing to replace the gasoline annually isn't that bad. I'm planning to put it in my car and being the cans to the gas station to refill. I have a reminder set and I expect it'll take me about an hour each time? I do think there's a good reason to store gasoline as well as propane, in case what you end up needing is fuel for the car.

3: In a bit of reading this may not be enough for something that is used as infrequently as I'm expecting to use the generator, because you do still have a bit of gas that doesn't get burnt even when you run it dry. This is what I was referring to with "doesn't require gasoline's level of cleanup from intermittent use". My current plan is to only run it on gas if I've already used all the propane.

1: Looking, I don't see people having problems with stabilized fuel after ~1y -- is this a problem you've seen?

Yes.  Shockingly high failures.  Out of, I dunno, 8 lawnmowers and 4 gas weed-eaters and 3 gas edgars in my life, basically 100% of them have either failed or were extremely difficult to start at one point or another from clogged carburetors and stale gas.

Yes, fuel stabilizer works, but it isn't perfect and you don't want to be trying to diagnose the issue when you actually need a generator.  It may be at night, or after the heat has failed during an ice storm, or during a rain storm, or when you're in a hurry to get your workstation back online because you're missing a critical meeting or demo.  

Note that a lot of people using fuel stabilizer are using it for cars.  They have much larger orifices, are harder to flood, have computer controlled starting systems etc.  The gas can be pretty mediocre and they still run - various videos on youtube people have tried substituting ethanol, etc, and the engine still starts and runs.

It sounds like maybe you're talking about small engines with stabilized fuel that are used very occasionally and aren't cleaned out between uses? I'm not talking about storing stabilized fuel inside the generator, but about keeping a gas can with stabilizer in it.

A generator is in that class of "small engines with stabilized fuel that are used very occasionally and aren't cleaned out between uses".

In order to clean it out you must disassemble the engine partially, simply running it to fuel starvation is insufficient.  Few have the time or tools to do this.

There is fuel still in the engine even after it sputters and dies (fuel starvation)

From experience, stabilized gas from a gas can still doesn't start the engine as easily as actually new gas from a gas can.

... There is fuel still in the engine even after it sputters and dies

Yes. If you look at our earlier conversation above you can see I'm also storing propane. I'm only planning on using the stored stabilized gas if the propane runs out (or if I need it for the car). Needing to clean the engine thoroughly before putting it away if I do use gas will be a project, but I think a reasonably interesting one.

stabilized gas from a gas can still doesn't start the engine as easily as actually new gas from a gas can

That's much more relevant for my plans, though it's hard to get a sense of how big a consideration this is.

It seems to me like a lot of the kind of people who want their own solar panels are also the kind of people who like the idea of being resilient in case of a blackout. 

In a competitive market, there should be an interest for a provider to advertise that they actually provide blackout resistance. 

The impulse to want to fix the perceived market failure with a government mandate seems to me a bad general heuristic. 

A better way to fix the issue would be to start a campaign of tweeting at Elon to complain that the solar provided by Tesla does not provide resilience in case of a blackout and that Elon should fix it. 

The impulse to want to fix the perceived market failure with a government mandate seems to me a bad general heuristic. 

OP says there was a 2014 government regulation that made it difficult to design & manufacture these systems, and then there was a 2017 government regulation that made it even more difficult to design & manufacture these systems—so difficult that there were no legally-compliant systems on the market until 2021. And if a company does go through all the effort to make a legally-compliant system, they have zero assurance that the electrical code won’t just get even more strict next year, and then all that effort was wasted. So I think it’s at least not entirely a market failure. I think Jeff’s proposal could help, and I think tweeting at Elon could help too, and maybe there are other approaches as well.

I do think there are also externalities involved—“societal resiliency during an extended blackout” is a public good. (If people are desperate and dying, they can cause problems for everyone else too.) So there’s a government-should-help argument from that direction. Also, the government subsidies already exist, it’s just a question of who qualifies for them.

The OP didn't refer to any government regulation. The National Electrical Code is created by a private trade organization, the National Fire Protection Association. Unsurprisingly, the National Fire Protection Association cares a lot more about protecting its members (a lot of them likely firemen) than about ends like resilient electricity production.

The increased regulatory burden is an important part of the Great Stagnation. The instinct to always want to fight problems by adding more regulation leads to stagnation even if you can find arguments that every regulation will help the public good.

The OP didn't refer to any government regulation.

As Steven says, the NEC is effectively a regulation since most states adopt it by reference (like the building code). If you want you could view my post as objecting to states adopting the 2014 and 2017 NEC without consideration of the effect on resiliency, instead of to the drafters of the NEC failing to consider this aspect? (I would prefer to have seen both.)

The instinct to always want to fight problems by adding more regulation leads to stagnation...

In this case I'm proposing adding a condition to a subsidy, which I think is an importantly less restrictive approach than a blanket regulation. People always have the escape hatch of skipping the subsidy if it's overall too burdensome.

One of the links in the OP says:

NEC 2020 brought more changes. The six simple sentences from 2014 have grown to a full page. Most of the addition is simply clarification of the requirements added in 2017. Since many states didn’t enforce the module level shutdown from 2017-2019, most people refer to 2020 as the release year that module level shutdown was added.

That does sound to me like some states decided it makes sense to not implement a part of the rules in 2017 because they considered the rules too burdensome/wanted to give companies more time.

https://www.renewableenergyworld.com/solar/evaluating-the-case-for-module-level-shutdown/ is an interesting article arguing that the rules that require the expensive module-level shutdown are essentially security theater.

many states didn’t enforce the module level shutdown from 2017-2019

I wonder whether this means that they added explicit exceptions, or that they just didn't adopt the new revision immediately?

https://up.codes/viewer/california/ca-electric-code-2022 seems to be the Californian version. In it they seem to have "adopt/adopt with amendment/adopt expect"

https://www.tradesmance.com/career-central/current-california-electrical-code also writes:

While there is a 2020 edition of the NEC, California — along with 23 other states — is still using the 2017 NEC. If you want to check out where we’re at, and what other states are doing, you can look at this color-coded map. It’s published by the National Fire Protection Association (NFPA), the same organization that issues each edition of the NEC. 

But, wait. Doesn’t the state have a more current electrical code? Yes, but the 2019 California Electrical Code (California Code of Regulations, Title 24, Part 3) is based on the 2017 NEC. 

In fact, we’re probably sticking to the 2017 NEC for a while. We didn’t vote to adopt this code until January 2019. The codes weren’t published by the state until July 2019. Then, the state observed a six-month statutorily required waiting period, which means the 2017 NEC didn’t technically get enforced until January 2020

Thanks! It seems like part of my beef is with Massachusetts being very eager to stay on the most current code. It looks like the 2020 NEC was adopted 12/27/2019, and the 2017 NEC 1/1/2017.

I hadn't realized how atypically fast MA is doing this. MA was the only state on the 2017 NEC from January until June; CO adopted in June, ID, ND, SD, MN, WY, and WA in July, OR and VT in October, and ME in November. Everyone else not until 2018 or later (or still hasn't).

Oh, thanks for clarifying. Indeed, NEC is not directly a government regulation. However, if essentially every part of the USA has a law that says “You must comply with NEC”, then NEC is effectively a government regulation.

(Some regions require NEC + amendments, but my impression is that the amendments tend to be more rules to follow on top of NEC, rather than allowing parts of NEC to be ignored. I could be wrong.)

It does seem problematic that a private trade organization, presumably with direct accountability only its own members, has the de facto ability to write laws that are binding on pretty much everybody. But I don’t know what to do about that. :-/

One of the links in the OP says:

NEC 2020 brought more changes. The six simple sentences from 2014 have grown to a full page. Most of the addition is simply clarification of the requirements added in 2017. Since many states didn’t enforce the module level shutdown from 2017-2019, most people refer to 2020 as the release year that module level shutdown was added.

That does sound to me like some states decided it makes sense to not implement a part of the rules in 2017 because they considered the rules too burdensome/wanted to give companies more time.

https://www.renewableenergyworld.com/solar/evaluating-the-case-for-module-level-shutdown/ is an interesting article arguing that the rules that require the expensive module-level shutdown are essentially security theater.