Hacker News new | comments | ask | show | jobs | submit login
The price of solar power just fell 50% in 16 months (electrek.co)
756 points by prostoalex on May 3, 2016 | hide | past | web | favorite | 435 comments

The aggressiveness of the price drops on solar is pretty amazing. I just lost one of the inverters on my grid-tie system after it had run trouble free for 13 years. So we're looking at the first "major" repair to the system I installed in 2003. And in 2003 my 5.2kW of panels on my roof cost $38,000 before subsidies and $19,000 after. 28 panels, two 2.5kW inverters, net about 4.2kW of generation.

I priced out replacing all the panels (now I can do in 10 panels what used to take 28) and the inverters with a single 5kW inverter, $7,000. My price, no subsidies. That is over 80% reduction in cost, over 13 years, of the list price of $40K.

And that is why it is a huge problem for the power companies, it makes less sense now to not have panels on your roof.

> And that is why it is a huge problem for the power companies

It is a problem, but not necessarily in the way you may be thinking. Peak residential power usage is in the evening when solar isn't available. Power companies therefore end up spinning down coal and natural gas during peak solar, then just spinning them back up again as the sun begins to set. [1]

So you haven't replaced the coal and natural gas plants, you're just running them less. This results in higher overall costs, because you pay $$$ to build huge plants then just let them sit most of the day, which results in higher per-watt cost of non-solar power because the infrastructure and maintenance is no longer amortized over whole-day usage.

[1] My friend, a manager of 30 years at a huge mid-west power company

Can validate. The problem with electricity is that there is no large scale storage like there is for other forms of energy like NG and Crude. Until someone comes along and figures out how to store electricity, this will continue to cause grief to power companies. Power plants are really just options, if the price per kWh is at a certain point then it makes sense to spin up or take down a plant and generate power. Adding a bunch of highly variant generation to the grid can make the market more volatile since there could be a lot of up and down with generation facilities.[0]

That said, there will be teething pains but this is a net positive thing that is happening. Growing up I lumped solar into the perpetual 10 years away category. It looks like I am going to see solar be competitive after all.

[0] I implement trading and risk management software in the energy sector, including Crude, NGL, NG, and electic.

Oncor solicited a study that showed that a mass install of Tesla Powerwall devices would be an economic solution to this problem [1], at which point all the generators lost their minds and lobbied the PUCT to prevent Oncor from destroying their market [2] (because of their position as a Transmission & Distribution Service Provider in the market, it would require a change in law for Oncor to be able to implement the project).

[0] Work for a multinational in their energy & utilities consulting practice; above comment is my own.

[1] http://www.brattle.com/system/news/pdfs/000/000/749/original...

[2] https://www.texastribune.org/2014/12/15/state-law-could-shor...

It's interesting to see the stark contrast between the naysayers that don't have information to back their claims (such as comments sibling to yours), and the people who have looked at current tech and made realistic assessments like Oncor and Brattle.

A lot of people are willing to make confident, but very wrong, statements based on out of date information. It serves to protect entrenched interests and stop competition, but I have a feeling that disruption is far closer than anybody is publicly saying.

X is realistic and smart because it agrees with my opinion

You're amazing

Thanks. But it's not realistic and smart because of that, it's realistic and smart because it examines the facts at hand and makes plans instead of handwaving. This is an important distinction.

Some countries store energy by pumping water up a dam. Seems efficient.

Very. One of the first was in Scotland. Great to visit as you drive into the mountain and I think one of the Bond files had a scene filmed in it


Nope. It was filmed in Switzerland. https://en.wikipedia.org/wiki/Contra_Dam

Different films. "The World Is Not Enough" had scenes filmed at Cruachan, "GoldenEye" was at Contra.

Whoa! That sounds really cool so I looked it up. It's called "Pumped-storage hydroelectricity"


Yeah, and I believe that it's the only game in town for storing really large amounts of electricity.

It's very efficient and it's a great solution overall; the only problem is there are only so many dams we can build, since each one requires drowning a valley.

Though you need less volume if it is only being used for overnight storage rather than long term rainfall storage. For example, to store enough energy to produce 1GW for 1 day (~10^5 seconds) with a head of 100m, use Pt=mgh to work out you need to store:

m = 10^9.10^5 / (10*100) = 10^11 kg of water. = 10^8 m^3

This is not a big reserviour, being 1/20th of the size of Warragamba dam's reserviour (in Sydney). One valley could provide 20GW for a day. Compare that with demand for where I live:


Peak demand for 7.5M people is 9GW, with an average closer to 7GW. One valley could provide overnight storage for over 20 million people, with first world demands (all of Australia). The limitation is transmission and generating/pumping capacity, rather than storage volume.

Note that's a winter peak you're looking at.

The highest summer peak appears to have been just above 13GW. http://www.wattclarity.com.au/2016/02/highest-electricity-de...

However your point still stands I believe.

But the summer peak is due to air conditioning during the day, when there's tons of "free" solar energy.

The more sun, the more water in the pool. Win-win!

Seriously though has anyone crunched the numbers on rainwater storage for microgeneration in a domestic setting.

Could I efficiently use excess solar to pump water in to a house roof storage area and generate electric from that?

They crunch the numbers here: http://physics.ucsd.edu/do-the-math/2011/11/pump-up-the-stor...

The energy in one gallon of gasoline is equivalent to the potential energy stored in 13 tons of water one kilometer up

The Tesla Powerwall has 6.5 kwh = 23.4 M joules.

Gravity potential energy = mgh. Suppose pool is 10 meters up, then you need 239,000 kg of water = 63K gallons or 8400 cu ft. This is a huge pool: 29 ft. x 29 ft x 10 ft.

You could use the city water supply: borrow water through a turbine at night, and return it during the day :-)

Or have pool up very high, for example on a cliff overlooking the Mediterranean:


The outflow from an Auckland homes storm water is not insignificant - an average roof would have something around 5 cubic metres of water come off it several times per year.

I remember a while ago on the show "Let's Get Inventing" (Saturday morning kids show with kids inventing stuff, think it was on TV2), they tried generating electricity from drainpipes, they couldn't get enough power to even power a servo squeezing the trigger on a spraybottle, they had to get a great big firehose from an firetruck to actually generate anything.

"Have you seen the swimming pool on the roof?"


Pumped storage is routinely used to store energy. It's not theoretical.

Has Flywheel storage every been seriously used in large-scale commercial applications? Someone mentions it on every energy-storage thread, and I'd like to know if the people suggesting it are just spinning our wheels.

FES is used in non-kill-zone habitable solutions like datacenters, bottoms of passenger trams/hybrid buses and electromagnetic aircraft catapults. The historical issue with FES is unscheduled, rapid disassembly, but carbon/kevlar composites have been shown to have greater integrity with least unsprung structural mass (because you want the most mass as close to the rim as possible to ensure highest moment of inertia.)

FES has 10x specific energy than supercapacitors or batteries with no memory and precise state of charge. Magnetic bearings and hard vacuums make them quite efficient.

The JET fusion reactor in Oxfordshire uses flywheels to provide very large pulses of electricity for short periods of time (a minute or two iirc). The reactor requires too much energy to run directly off the National Grid.

As I recall from a tour I went on of jet 30 years ago there are a large bank of flywheels which spin up over several minutes, these are then rapidly discharged to a bank of capacitors which are then used to power the experiment, I believe the duty cycle was 9 minutes. [1][2]

I believe some electric cars used capacitors to store regenerated electricity from breaking.

"More Ampères please Mister Woodbine." - Road to Welville

[1] - http://pia.sagepub.com/content/200/2/95.abstract

[2] - http://www.euro-fusionscipub.org/wp-content/uploads/2014/11/...

Sounds like they use it for high output rather than efficiency though. Still interesting that it has real world applications though. Shows that the technology has potential as well.

It's not even remotely on the same scale. There are three massive utility-level flywheel deployments according to Wikipedia that provide 15 minutes of output.

Edit: Seems the purpose is purely frequency regulation, rather than storage per se.

Better than nothing, but not that efficient. I took the tour of the somewhat mis-named Dinorwig Power Station (really a pumped storage facility) in Snowdonia, Wales. The figures they quoted there put it at about 75% efficiency. It's not really supposed to be for 24-hour power smoothing, more for rapid response to sudden demand; it's a lot faster to open a valve on a turbine than to stoke up a coal plant, and for some spikes (such as the example they quoted on the tour, when everyone puts the kettle on when Coronation Street finishes) you have a 30-second window or so. However, as demand has increased it's being used more and more, and the more it's used the more power we're wasting.

Interestingly, it does not use a dam. There's a natural lake at the bottom of a mountain and a man-made reservoir at the top, and the entire pumping facility is completely inside the mountain; it's not at all obvious that there's anything there at all from the outside. I think about that whenever people moan about the environmental impact, or flooding valleys.

Works great if you have the right geography.

I thought I read that some water districts run their pumps at reduced volume during peak hours and top them off before or after. It's a smaller volume but it's already a solved problem.

And it allows you to push some levers and dials as to whether it's cheaper this year to increase your water capacity or your power capacity. City planners love that kind of thing.

I grew up by one of these in NJ: https://en.wikipedia.org/wiki/Yards_Creek_Generating_Station

I remember parking at the lower reservoir and hiking to the upper one and being amazed at how big the water pipe is.

It's a problem, but the nature of the problem varies a lot from country to country. In the Netherlands, we don't have enough renewables yet for this to be an issue (~10%). Once you start getting to Germany's penetration, you start to see some issues with balancing the grid and meeting peak demand. There are more options than just storage, though.

Demand side response is already used in industry to regulate electricity demand and smooth peaks. Smart meter penetration will allow utilities to do this more and more in the residential sector too.

Supergrids are also starting to emerge and indeed, some large grid companies are pushing hard for international interconnects. (China wants a supergrid). Countries are building out more and more HVDC interconnects, both across borders and within countries. Once you have a network of utility scale renewable sources across multiple countries, meeting peak demand becomes more a matter of coordination between grids. Europe for example, is looking like a very interesting area for this right now with planned interconnects between North Africa and Southern Europe.

Since we're all citing employment, I work for a renewable energy company :)

Instead (or rather in addition) to storing power, we can also get better at coordinating demand.

There are already programs for industrial users to get electricity for much cheaper, if they can tolerate power cuts during peak demand. Those programs will become more and more rewarding to join and extend and automate as supply gets spikier.

That's great but then instead of increasing the effective capital cost of the generating plants, you're increasing the effective capital cost of the industrial facilities (by lengthening the time to pay off financing).

It'd be interesting to see a study comparing the two options.

The incentives create a market in which the energy consumer and producer can trade in their existing inefficiencies. The market will decide where it is best to make the compromise, whereas before there were just wasteful externalities.

If you have an industrial process that can cope with intermittent power supply, it may be only slightly more expensive to design or build so that interruption of power won't result in interruption of production. If the overall cost increase is less than the decrease of energy costs due to incentives, there is no downside.

The energy producers likewise price the incentives so that their loss of revenue is lower than their cost savings.

Unless of course the market is created and operated by Enron, then we're all fucked.

Yup. Having many distributed li-ion battery walls charging at night for peak assist during the day is one way to level out demand and make the energy infrastructure more resilient in a Google-servers-like way.

Is there a way to coordinate demand without asking thousands of people to give Alice the ability to shut off their appliances?

Sell electricity at spot prices, give consumers access to that spot price, and you'll see a lot of demand flatten simply through natural market forces.

Where I live, I pay five times more for electricity between peak hours of 2pm and 8pm than I do between off-peak hours of 10pm-8am. (The remaining hours are priced at a medium level.) This has changed my behaviour, in that I now tend to wait until bedtime before switching on the clothes dryer and dishwasher.

No, although the current system is pretty tame. The electric company can turn your electric water heater or air conditioner (depending on geography and utility) off for 15 minutes at a time, and in return you get a credit on your electric bill. Its opt in.

Actually, there already is, it's called OpenADR.


See also Ecobee vs. Nest: http://controltrends.org/controltrends-news/news-and-informa...

Yes, industrial plants will turn off production (for instance at an aluminum smelting plant) in relation to the frequency of the grid. Grid frequency decreases slightly when demand starts outpacing "supply."

That frequency change used to be a natural reaction from the generators.

I wonder whether they create that signal artificially with electronics these days, or if it's still a natural consequence of the mechanics of the spinning generators they run in power stations?

I recall reading a while ago about using molten salt storage at large solar concentrators. Did this ever take off? The theory was you generate excess power during the day and you can use this excess to heat the salt solution which then release thermal energy during the night.

Does anyone know more about this? Is this able to meet baseload demand?

This one in Nevada came online not too long ago and is quite large -- apparently has 10 hours of storage and according to Wikipedia it generated 9.1GWh in February 2016 alone.


Solana Generating Station, a 280MW solar concentrating plant, currently, uses molten salt, which allows for six hours of energy storage.


Thermodispatchable solar, even if the word does not exist, has been a recurring theme. But installations seem to be one-offs that are rarely followed by a direct successor project with the ever dropping price of photovoltaic.

In areas with cold weather, one of the most interesting (and underrated, due to only making fossil use more adaptable) developments is the installation of huge, insulated hot water tanks, to make the power generation and the great generation of combined cycle plants individually dispatchable.

Sorry for the late correction: in the last sentence I meant cogeneration plants, not combined cycle. (which would typically not be combined cycle, because the heat not captured for electricity is not wasted in cogeneration)

http://www.lightsail.com/ has been working on this for a while now

We have a lot of good technologies already, the problem is they just don't scale well.

I can envision older warehouses in an industrial district being converted in to vacuum flywheel storage sites where power fed in to the grid earlier is buffered and then released back to it.

Changing our culture is another possible approach. Power is used in the evenings because that's when most people get off work/home/etc and they tend to have /that/ time to do cooking and cleaning (thermal cooling / heating are more solvable with changes in materials and building shape than culture change).

Price electricity differently through the day according to its scarcity and people might eventually adjust their routines to reduce evening usage.

That seems backwards. The aggregate preferences and habits of humans should be seen as an incentivizing force for the creations of new technologies, not the other way around.

Yikes. Yes, of course technology is driven by human preferences. But, No, humans must occasionally accept that certain habits are not sustainable given current and near-future technologies.

I generally agree with your position but in this particular scenario the people with the greatest capacity to adjust their behavior are the people who will be affected the least financially by a hike in peak prices and thus less likely to change or more likely to invest in solar at home, while those with the least flexibility in work arrangements and spending, e.g. lower income earners working in the services sector, will be affected negatively the most and unable to work around it.

If the spot market was the status quo, and someone was suggesting a change to the current system of flat prices, one could make the same argument. (Ie already well off people are flexible enough to make use of the new system better.)

More variant pricing will better distribute usage. The unfortunate effects on the poor can be negated by a sales tax paid back to all citizens equally as a lump sum.

Well, in that case, pricing electricity differently through the day will encourage people to develop new technologies for cheaply generating and/or storing electricity for evening consumption peak.

That's like saying market economies are backwards. We need to match resources to people somehow, and markets are a very popular way to do that fairly and efficiently.

I'm not saying that markets should be unchecked but it's very strange to see such naked Marxism on this site.

I don't see any Marxism there. Marxism isn't just anti-market. It's a complex ideology.

It doesn't have to incentivize new technologies. It could just as easily incentivize new business models. Imagine a laundromat that sells "FREE DRYING" during the day. Heck, I bet you a lot more people will do laundry during the day time. Similarly, power companies can give incentives for more day time power usage. A lot of people already follow night time power usage today so its no reason to think it wouldn't work.

Variable pricing would create those exact incentives.

It's a two way street.

Aggregating the preferences via price will drive new technology and business models.

You are correct- there are energy storage solutions out there, I should have clarified my statement above. The issue is that they cant store power grid levels of power in an efficient, compact, and scalable manner. This is the rub. Maintaining a giant battery bank or other measures require maintenance and large capital expenditures.

Being a manager at a large power company for 30 years is both an advantage and a disadvantage. For the past 30 years, the energy industry has changed incredibly slowly, so ones expectations of what is possible in the way of innovation is far far lower than it should be.

The "duck curve" that your friend is referring to will be solved in a variety of ways, from wind to storage to HVDC interconnects to the rest of the nation, to peaker plants.

In the future, coal is dead. No need to build any more of it. Nuclear is too expensive to build anymore. (Unless somebody actually does thorium innovation, but I'm not holding my breath...)

When storing a kWh costs $0.10, and there's also adaptive pricing to match supply and demand better, these problems will melt away.

It does require a huge mental shift, but the cost savings will make that happen

> When storing a kWh costs $0.10, and there's also adaptive pricing to match supply and demand better, these problems will melt away.

If the math and assumptions in this article are correct, then that's exactly the (wholesale) price point Tesla is aiming for with the Powerwall:


There are a couple other (not yet released) energy storage solutions mentioned there that might get the cost down even lower - Eos Aurora and Imergy Flow Battery

It'd take over 1 BILLION Powerwalls to store half the daily usage in the US (using old usage numbers off Wikipedia).

I wonder if there's enough raw materials to make that realistic. And how to you manage recycling, repair, replacement, end-of-life concerns for all those batteries on that scale?

Not asking a sarcastic hypothetical. Genuinely curious. It's hard to wrap my head around numbers that big.

For reference I used this: https://en.wikipedia.org/wiki/List_of_countries_by_electrici..., divided it by days, then the Powerwall's 6.4kwh capacity and then halved that number.

4,686,400,000,000 / 365 / 6.4 / 2 == 1,003,082,191.8 Powerwalls.

Just throwing around "BILLION" doesn't make it seem big, you need to compare it the unit size.

A single gigafactory is going to produce 35 GWh in a year, which is about 5.3 million power walls a year. If their lifespan is 10 years, than we need to produce 1e9 power walls / 10 years = 100 million powerwalls/year to maintain that capacity.

So we'd only need 20 gigafactories worth to do this. And with wind and proper grid interconnects we won't even really need much storage at all to keep the electricity going; the national weather is very very consistent.

We will need enough lithium for our ~300 million cars and their batteries, which are probably going to be 50-100kWh each, which is far far more than storing half a days worth of electricity.

What if we used tens of millions of electric vehicles on the road as grid storage?

Tesla has already planned on it: https://i.imgur.com/oLBNeW4.png | https://i.imgur.com/pectzXn.png

That's over 3 Powerwalls every second. And I've got to think you're not going to see more than 75% original capacity in ten years as a best case?

I'm not as concerned about the factory as the supply chain, can we actually supply that much raw material, and what will mining it do to the environment? I assume lithium is much rarer than iron or copper. But I don't really know.

For reference 16 million vehicles were sold in the US in 2013, but many (most?) of those were trucks and large SUVs, which AFAIK aren't on the roadmap for any EV manufacturer right now. And then there's fleet/work vehicles, single car families in states without long distance transportation or quick chargers (like TX).

It makes me wonder if 20 years from now fuel cells won't completely obsolete batteries?

Residential is only about 1/3 of consumption. I guess at least industrial won't be using Powerwalls to store backup power.


It might not be possible to make all those batteries using lithium, but that doesn't really matter for stationary batteries.

That's a huge factor I hadn't considered. Thanks.

Raw material supply is a big issue; see Berkeley physicist Tom Murphy's calculations: http://physics.ucsd.edu/do-the-math/2011/08/nation-sized-bat...

He's mainly talking about lead but for lithium it's even worse.

A lot of things look great until you look at the scaling limits. Hydro storage is another example; Murphy's got a post on that too.

If these numbers are even close to accurate, this should be the top post.

It's no use for everyone to argue over wether a completely implausible idea is financially competitive on a small scale when the implication is we're talking about scaling on a national level.

The number actually needed is a bit lower than that (I estimate half a billion), but the general question still stands.

Variable electricity needs already mean we use less electricity at night time [0]. Shift electricity expensive processes preferentially run at night to the day and using 80%/120% seems like a reasonable rough approximation. So we need 4,686,400,000,000 / 365 / 2 * 0.8 ~= 513,5780,822 kwh in the '12 dark hours'

Between current supplies nuclear hydro and wind we account for approximately 30% of electricity generation [1], on the assumption that electricity generation from these is constant, but can't be ramped up during the night, we get about 4,686,400,000,000 / 365 / 2 * 0.3 = 1,925,917,808 kwh. Leaving 3,209,863,014 kwh needed from storage, or 501,541,095 powerwalls.

[0] http://www.eia.gov/todayinenergy/detail.cfm?id=13131

[1] http://www.eia.gov/tools/faqs/faq.cfm?id=427&t=3

Think about if Tesla vehicles were used instead of Powerwalls: https://i.imgur.com/oLBNeW4.png | https://i.imgur.com/pectzXn.png

Your car isn't at the house while solar is being produced though for many/most people right? This seems unrealistic.

I guess it could drive itself home after dropping its owner off at work, but that's probably too inefficient in most cases to make a full extra round-trip just to charge up with the panels at home.

Its probably sitting at your employer's lot. A lot that that would have charge stations.

None that I've worked at. And aren't small businesses the biggest category of employer?

I guess toomuchtodo is talking about the future, not the present.

I am!

A lot of cars sit in carparks during the day.

I guess the implication being there'd be a massive number of charge spots and you could swipe your card to pull off the grid what your home is putting on.

But this would mean turning a car-park with just 40 plugs into needing 1,600 amps at 240v. With those grid connected homes having modest 2kWh solar installations you're talking 80kw/h. If you only spent 3kWh getting to work on average (double that for the round-trip) you'd fully charge the cars in a couple hours.

So I guess the goal would be to run your home off the unused capacity. But then you're putting a far heavier duty cycle on those batteries. Whose going to pay for them when that $10,000 pack needs to be refurbished in five years? Or do we just accept EVs depreciating down to zero when the pack replacement cost exceeds their value?

There may well be good answers to all this. Just some things that popped into my head.

It would seem at first glance to be very costly for the car park, the grid and the consumer. I'm not sure any technology with the depreciation schedule of lithium batteries makes sense financially or environmentally?

Thanks for working through some numbers.

My modest intended implication was that if we only care about total sums, where the car is doesn't matter too much, as long as it has access to the grid.

The economics could work out, if the car park charges you more for electricity than they pay the grid; and the grid your home less that it charges the car park.

The car park and the grid both take a cut. For the home- and car owner, this works out to a slightly less efficient battery cycle. (If you convert the money charged back into the equivalent amount of energy.)

Of course, if you are going to use your electric vehicle like as a battery like this, it's going to affect the battery's useful life. The same would happen with the single purpose battery you have sitting in your home.

Now, the performance characteristics one looks for in a car battery and a grid storage battery are different. So I don't know whether using the car battery as a home battery even makes sense. (Ie home batteries can be bulky, but better be cheap. Car batteries have to be light and charge fast.)

On a more refined note, one can play the same game, but only do it during spikes. Ie the car can stop charging during a spike in demand, and even uncharge if the spike gets big enough or the battery is already full.

This won't do much for the diurnal cycle (that we discussed above), but a contribution during the spikiest peaks might already be worthwhile.

My gut feeling is that we will see the demand shaping for charging, but I am more doubtful about whether we'll see cars actively feeding back into the grid.

> Being a manager at a large power company for 30 years is both an advantage and a disadvantage.

He is definitely biased against solar, but is acutely aware of the costs it presents to traditional power generation and definitely well qualified to comment on them.

Right, and my point is kind of that the entrenched energy interests have been wrong time and again about solar, wind, and now storage. As well as HVDC.

They don't understand these things, they see less than half of the equation because they don't want things to change, and they only see the problems and have no interest in solving them.

IT wasn't that long ago that they were saying that solar was impossible, or that wind would never be affordable.

They were wrong about that and they will be wrong about how difficult it is going to be to adapt to lots of intermittant renewable energy powering the grid. They will complain vociferously, and they may be able to talk about some of the difficulties, but we can't trust their perspective on the big picture.

> In the future, coal is dead.

What about coking coal, for the production of steel from iron ore?

I've heard the claim it's indespensible but that's not true from a chemical point of view. But possibly from an economic point of view.

Nuclear could be made cheaper (even with Uranium) by updating regulations.

Fission plants are held to a much higher standards of safety in terms of eg decrease in quality adjusted life-years per Gigawatthour than any other form of producing electricity. Even solar.

(Though most deaths and injuries from solar come from people slipping off their roof when installing residential capacity. If memory serves right.)

A nuclear plant today would cost something like $6B that's not a regulatory tweak.

The problem with nuclear is waste disposal (still not figured out, and a will be a problem for decades/centuries after the plant is closed) and amortization of billions of capital costs.

With the price disruption potential of other technology, it's a bad bet, and has been so for many years.

Nuclear was only ever "viable" for political reasons. We had to have enough nuclear weapons to extinguish mammalian life nineteen times over. Eighteen times would have been pathetic. It was all a boondoggle anyway, so the war pigs were happy to spread a bit of cash on the civilian side...

There are nuclear plants that generate isotopes for medical purposes.

Those are not the same plants that generate power for the grid.

The problem is they have a huge spinup and spindown cost. So as long as we have solar, we need power that we can use at those times when solar is unavailable.

This claim about peak power usage is untrue, at least for California. For the past 20 years in California, the annual peak electricity usage has ALWAYS occurred between 2pm and 5pm.:


I used to work at a power company as well.

(It's true that the daily peak is in the evening and that won't be helped by solar. But capacity is built to handle the annual peak, not the daily peak, and because the annual peak tends to be hot afternoons, solar actually will replace other generation in the long run.)

"Lets not move forward until every conceivable issue is solved" - said every entrenched interest ever.

There is another way - minute by minute pricing. Much of the electric power usage is not required to be right now, and can be deferred - running the refrigerator, A/C, hotwater heater, dishwasher, charging the car, etc. With real time spot pricing information available, the deferrable usages can be deferred until the spot price is cheap.

I bet this could even out demand quite a bit.

With the current flat rate pricing structure, there is no incentive whatsoever to shift demand to blunt spikes.

Which is exactly why Tesla and other companies are investing in residential power banks. They have the potential for huge efficiency gains across the grid.

The major challenge now is no longer PV and gridtie inverters, but energy storage. Whether it's batteries, pumped storage hydroelectric, whatever.

$$$ per kWh stored (and the lifetime/cycle life + operating costs of that storage) is a key benchmark.

I am clearly out of my depth here but what if we take the newly daytime unused/idle coal power or natural gas power for other endeavors like sea water desalination or just pumping up water into a watertower when daytime energy cost is low and adding increased hydro capacity in the evening when the energy cost is higher?

Those sources of power have a higher marginal cost and are only spun up when the spot price is high enough - in peak demand times.

What you are talking about is demand shaping.

It's already happening, but I predict we'll see more of that in the future.

Eg lots of industrial applications like smelting aluminium can do with short-ish power cuts. In return, they get a steep rebate on their electricity bill.

> Peak residential power usage is in the evening when solar isn't available. Power companies therefore end up spinning down coal and natural gas during peak solar, then just spinning them back up again as the sun begins to set.

Would it be possible (in an utopian world) to actually have a global power network and route electricity produced during the day in one area across multiple timezones to an area where the demand is greater and no available solar? e.g. Electricity produced in West Africa at noon to be delivered to .. Tokyo, where it would be 9PM.

I understand that politically this is almost impossible, but I am curious whether there are any technological impediments for this to be feasible.

I would propose a global power grid as a solution to this problem, on which the sun would never set. The only problem is that earth is not set up well for this: the Sahara would be the solar workhorse of the planet, and the consumer of all this power would be Hawaii.

What if the grid connects (residential and commercial) across time zones, one generating for the next with no need of storage devices?

Maybe if we had room-temperature superconductors. For now, transporting huge amounts of electric power across many thousands of miles is very, very inefficient.

HVDC lines work very well even if they are quite expensive.

Paging Buckminster Fuller.

Panels islands.

Wait 5 more years and batteries will become cheap enough to even remove that constraint

> Peak residential power usage is in the evening when solar isn't available.

Assuming it's not a battery backed system, prices on those are dropping (or more correctly remaining stable while capacity slowly climbs).

Dropping, but still incredibly expensive compared to the alternatives =( ~doubles the cost of solar.

Well, you could just massively overproduce solar, and use the excess to pump water up a hill. Also, I think there are certain externalities with coal that your friend never had to pay for, like Mercury poisoning. Reducing the amount of time coal plants run may be beneficial to reduce those externalities.

That's a good thing. Peak power usage is midday, so providing additional supply at that time avoids the need to site new plants, which costs millions or billions of dollars.

It also makes coal less economical and makes gas plants, which are cleaner and safer to operate more economical.

Peak load is actually around 6pm.

I guess it depends on country.

Some countries need AC, others need heating.

The power companies can get overflow from solar and wind and water and redistribute it at a profit while paying the residential over suppliers. It's on the power companies to build the infrastructure like batteries to take advantage of all this free new power.

This is why we need a major leap in battery tech to make solar more viable.

Plus there is the problems of cities with not enough sunshine and otherwise problematic weather (snow on the panels).

Sorry for the noob question, but isn't there always sun somewhere on the planet? Are electricity grids not globally connected, or would it be lost in transmission?

They are not globally connected. And losses become significant after a thousand kms or so. You're not going to get to the electricity to the other side of the world where the sun is shining.

You need local storage. As much as is said about batteries, the only real solution here is hydro. Expect to see every viable valley start to get dammed over the next decade or so.

>Expect to see every viable valley start to get dammed over the next decade or so.

Except that dams are instead being removed for a variety of environmental and associated reasons. Free lunches are rare.


Is this DIY, or installed? If it's not DIY, mind passing along who you're using? Inquiring TX minds would like to know!

My wife and I designed the system but it was installed by a Los Gatos company that did not survive. As part of our education efforts, my wife completed a program to become a certified solar installer (you needed to be certified to get the full rebate at the time) but one of the women she met in the class was working at this company and it just made sense to give them the job to help them bootstrapping their company.

So what you're suggesting is to take the installer course and then hire one of your classmates.

Not really, I was just relating how it worked out for us. There are a number of Solar Power advocacy groups around, in the Bay Area we went to meetups of NorCal Solar[1] (back in the early 2000's they would tours where you could see houses that had solar and ask questions). Great way to hang out with like minded individuals and collect folks experience with various installation companies and system options. Installations are common enough these days that you can probably find someone who has already done what you are thinking of doing, and tell you how it worked out for them.

[1] http://www.norcalsolar.org/

How long until you start seeing a return on investment? Right now my electricity bill for my home is ~100 a month in Texas. I would need to keep the same panels without maintenance fees for 33.33 years by your price point of 40k

That is a slightly more nuanced question than you might think.

First, I work from a baseline cost of $19K, even though the list price was closer to $40K I didn't pay that for the system so I use the actual cost in my calculations.

Second there is the point where the money you've saved is equal to the money you paid, we hit that point about 10 years in. And then, depending on the life of your system (which has "positive cashflow" at that point) determines your overall rate of return. The replacment inverter (which is actually going to replace both inverters) is $2K ("upgrade" pricing from the manufacturer) And as the inverter was failing our system was under performing (so not generating as much power). The question will be how much longer am I willing to run the panels. The warranty is 25 years, so in theory another 10 years but the manufacturer (Sharp) no longer makes them and cannot replace them under warranty (they can only offer to repay you to buy a panel from some other vendor with equivalent specs and size).

So from a money perspective, had I used the $19K and put it into an S&P 500 ETF in 2003, it would have doubled by now (SPX was 1,008 in 2003, its over 2,016 today). So "bad investment" comparatively. If the panels continue for an additional 8 years and PG&E's costs stay about the same then I'll have generated about $20K in value from them so that would pencil out to a ARR of about 3.5%. Initially I funded the panels with a refinance of the house loan and that was at 6% but that has since been refinanced even lower. However had it not been, at my original purchase I'd still be running a deficit trying to both pay back the house and generate cash from the solar.

Interestingly for me, with the $7K it would cost to re-implement that system today, and 4% mortgage rates, it would have a rate of return of closer to 14% (10% if you subtract 4% as the cost of capital for a mortgage refi) and pay for itself in a bit more than 3 years and have doubled the return in 6. The challenge is additional regulatory changes which get incorporated into new installations versus the regulation regime I operate under based on my 2003 install date. I wonder sometimes what the trigger is for resetting the project's origin date.

I think you're being a little unfair with the S&P comparison: you should be comparing the returns on your actual investment with the expected returns on the S&P, not its actual returns. The stock market got a 5.4% return over that time, yes, but that's above average; we should expect its long-run returns to be close to the rate of economic growth, or more like 3%. And it could easily have been zero or even negative over a short time like 13 years.

By comparison, the risk that your solar panels will stop producing electricity, or that someone will invent Mr. Fusion and drop the price of electricity through the floor in the next decade, is quite low.

True, but its too easy to get lost in financial modeling. There were two high order bits in my decision, first was that the money was spent either way; Either I would be paying PG&E month to month, or I'd be paying off the system. And there was a good shot that the panels would generate as much or more energy for that investment. The second was that I felt it was the right thing to do, not a lot of people were in a position at the time to pay those sorts of costs and I really believe that finding better ways to power the world was a noble cause, by investing in a solar system, even if it was a 'bad' investment, it was jumpstarting the market for solar. That helped with the chicken/egg problem of people investing in new solar tech to meet a market that was (by the action of early adopters) clearly emerging.

So every time I find myself having this discussion in a group, there are folks who want to argue the financial model to "show" something (generally negative) about solar power. Either it's not a good investment, or it's just coal power used to make the stuff that now is generating not even as much power was was used to manufacture it, or it's causing more pollution by forcing utilities to turn plants on and off, etc etc.

I was fortunate that I could finance it without risking things like my kid's college fund, I live in an area where there is a lot of sun and mild weather so my house's "demand" is quite manageable, and as an engineer I get to live my belief that we can engineer solutions to the worlds problems like energy interdependence and climate change.

I am grateful that today, the systems costs have gotten so low that it's possible for more people to make that choice.

Just to be clear, I think it was a decent investment then (better than the stock market) and an excellent one now. I'm not one of the folks trying to show something negative about solar power.

I find your belief inspiring.

Dont forget your stock gains would have been taxed

May I ask where you got your 3% because that number is very low compared to any number I have ever read? This Wikipedia chart shows annualized returns to be 10.47%. That is >3x your 3%. https://en.wikipedia.org/wiki/S%26P_500_Index#Annual_returns

3% is an estimate of the rate of economic growth: http://www.tradingeconomics.com/united-states/gdp-growth As I said above, we should expect the market's long-run returns to be close to the rate of economic growth.

It's true that stocks in the US in the 20th century dramatically outperformed that 3%. This is for three reasons: first, up to about 1975, the world's economic growth rate was higher than 3%, due to the Second Industrial Revolution; second, up to about 1975, the US's economic growth rate was even higher than the world's, because it was taking over the markets previously supplied by the bombed-out economies of Europe (especially the UK) and Japan. Third, since about 1970, the share of economic output that went to owners of capital (rather than labor) increased dramatically.

Now, the third of those things probably can't happen again, because capital's share of output is already super high. (It's not that it can't get higher, but it can't go past 100%.) The second could happen again, but if it does, the US will be in the position of the UK, Japan, or Germany — it's now the Single Superpower (not even a mere Great Power) who could lose market share, probably to China or to a non-nation-state power. Investors in Japan's and Germany's stock markets in 1925 or 1935 didn't do so well.

The first, well, that's anyone's guess. It's a total wildcard.

It's certainly reasonable to posit a Third Industrial Revolution, made out of some of free software, nanobots, biotech, abundant solar energy, AI, and automated fabrication. I sure hope we get one. But if we do, it's not at all obvious who the returns will flow to. (Everyone, I hope, not just shareholders.) Accumulating capital goods, as we've been doing for three million years, is less important when your capital stock can double every 24 hours through self-replication. Breweries do not account for the quantity of yeast they have on hand as a durable capital asset.

And it seems equally plausible that we'll instead experience a new Bronze Age Collapse or Decline and Fall, as inexpensive DIY drones, very affordable precision projectiles, anonymous markets in assassination, and ubiquitous retail surveillance provide a decisive advantage for attackers over defenders in the physical world, just as their software counterparts have on the internet.

Returns to the S&P should beat GDP growth even in the long run, steady-state.

GDP growth flows to both debt and equity. The debt market is actually much larger than the equity market and since its returns are lower than equities, and they both share GDP as a source of returns in the long run, equities should beat GDP growth in the long run. This argument isn't iron clad, but hopefully you can see the general picture.

The Gordon growth formula is taught in intro to finance classes to estimate returns to equity. The theory is that stock market returns equal the dividend yield plus growth in stock prices. Stock prices alone are in the long run expected to grow at the same rate as GDP, but the actual return should be higher by the rate of the dividend yield.

These are the most basic arguments against what you're saying, but I would also caution against assuming that any particular macroeconomic trend will end, especially that it will end in a timeframe relevant to decisions related to solar panel installations. Sometimes they just keep going. There is no mathematical reason that equity prices can't go to infinity, as required rates of return can decrease to zero. If the world becomes more predictable and stable, equities can keep becoming more and more valuable with no damage to finance theory.

There's not much reason (and even less empirical evidence[1]) to believe that U.S. stocks should follow U.S. economic growth rates.

[1]: http://www.economist.com/blogs/buttonwood/2014/02/growth-and...

To summarize the article, the long-run correlation between stock prices and economic growth rates is positive at 0.51, and in the short run the correlation is actually negative (no number given), because investors bid up stock prices in fast-growing economies, because they expect the growth to continue.

If the growth did continue, then they would make money, but often it doesn't, so they overpaid, so they get below-average (and often below-break-even) returns.

Also, per-capita GDP growth is irrelevant; what correlates positively with equity returns is total GDP growth, including the change in population.

So, on the contrary, this article provides a great deal of reason and even empirical evidence to believe that returns on US stocks should follow US economic growth rates.

I understand the rate of GDP growth, the person you replied to specifically stated that they would have made more money on the SPX. Granted the S&P 500 is not guaranteed to go up, but historically it does.

"...first, up to about 1975, the world's economic growth rate was higher than 3%, due to the Second Industrial Revolution; second, up to about 1975, the US's economic growth rate was higher than the world's, because it was taking over the markets previously supplied by the bombed-out economies of Europe (especially the UK) and Japan."

Do you read the posts you reply to? Because the chart I linked starts in 1970. And you are discussing GDP when the post you were replying to specifically was talking about investing in the market.

I'm not an expert here, but doesn't it make sense for stocks to return higher than growth because of earnings?

For example imagine I own stocks making up 1% of in a company w/ earnings of $X, and market cap of 10 * $X. If the companies growth over a year is 3%, at the beginning of the year my stocks should be worth 0.01 * $X and at the end of the year 1.03 * 0.01 * $X due to the growth, BUT shouldn't i also have received 0.01 * $X (my share of earnings) in dividends that I am free to reinvest, giving me a 4% return overall?

If the company's value increases by 4% over the year, and it pays out one of those four percentage points as a dividend and reinvests the other three, you get the scenario you're talking about. (And it doesn't matter how much of that 4% is revenue minus cost of sales and how much is, say, increase in value of its holdings in another company or real estate.) But if the economy grew by 3% and the company's value increased by 4%, your company is doing better than the economy as a whole, and it's because someone else is doing worse. In the US over the last 40 years, a very significant "someone else" here has been the employees — as they have lost bargaining power, they have gotten worse bargains.

It seems like maybe you're saying that the revenues, and therefore the profits, don't count here because they're not part of economic growth, but just part of the ongoing economic activity. But what determines the P/E ratio, which is to say the return on capital (as a reciprocal), across the market? What keeps investors from bidding the market cap of the company up from ten times earnings to a hundred, a thousand, or ten thousand times earnings? It's the availability of other investments that they expect to grow in value at a higher (risk-adjusted) rate. If you can get a 16% annual return on your investment by buying solar panels instead of stocks, then the guy who does that will have 16% more money to invest in more solar panels every year, until either he bids the price of solar panels up (due to limited manufacturing capacity) or he bids the price of electricity down (due to limited transmission grids or electrical demand). The first of those is already happening; the second one should start happening in about 2024, earlier in some areas.

Now, you could argue that there's a difference between this solar panel maniac guy spending 16% of his capital base in solar panels every year, accumulating more and more solar panels and selling the electricity from them to buy more, and GDP growth, because the solar panel maniac is accumulating a stock, while what the GDP measures is a flow.

But note that by the hypothesis that the maniac is investing to get some relatively inflexible percentage return on his investment, he receives a flow of earnings that is proportional to that investment. And that flow is part of the GDP.

GDP growth has no strict correlation to stock market returns. Why not? Because corporate profits (eg the S&P 500) can grow far faster than GDP, as they did from 2002-2014.

You should not expect stock market returns to mirror GDP growth. The S&P 500 is not strictly representative of US economic performance, which is precisely why the huge corporations did so well with the cheap dollar and significant global economic expansion while domestic locked companies didn't fair nearly so well.

Your point that the S&P 500 represents large US companies, rather than all US companies, is well taken. Still, the process of centralization you describe (where most gains go to past winners) can't continue indefinitely; the S&P 500 companies (at least the ones in the US) can't account for more than 100% of US production.

Additionally, you'd have to subtract the monthly power bill from those returns. I.e. you can't say "I would have doubled my $20k return by investing it for 10 years", because you can't defer your monthly power bill for 10 years and pay it in a lump sum after you doubled your money in the S&P. Paying $100+/month for power would substantially impede the interest compounding effect.

No, that would amount to double-counting the returns. Yes, you need to time-discount future returns, and you may need to risk-adjust, but that's already taken into account.

The rate of economic growth has little relation to the stock market yield. For one you have forgotten about the dividends, currently around 2%. Now lets assume the company grows by 3%, so your shares will now be worth 3% more. But the total return will actually be 5%.

However here I assumed that there was no drop in optimism, or pessimism over the one year period. For if you were to bought in when expectations were high, your actual return would be lower.

Sometime the relation between growth and return is completely paradoxical. You would expect that in the past 100 years a portofolio invested in UK stocks to be dominated by one of US stocks. After all the first country went from being The Global Superpower to dubious second rate global player, meanwhile the US had done the same in reverse. Yet an investor in UK stocks would have been marginally better of.

Likewise you would expect China with its spectacular growth, to have brought impressive returns compared to the US. Yet that had not happened.

You aren't the only one to make those mistakes; I've explained where you went wrong in https://news.ycombinator.com/item?id=11623817 and https://news.ycombinator.com/item?id=11623377.

Actually the markets are "supposed" to double every 10 years, which is said that on average over 10 years, the markets will grow 10% per year. Since the recession 8 years ago that has been much lower, but in general, that is the rule most industry people will give you. I work as a tech guy in finance and have for the past 8.5ish years.

The U.S. markets reflect the U.S. ascending to become the dominant world power over the course of a century. Naturally, the markets grew during this period of tremendous economic growth. The investment industry likes to extrapolate from this, and declare it a rule of the markets as if it's a law of physics.

If growth slows or stalls, eventually the markets will catch up. A market that grows independently of the underlying economy doesn't make sense in the long term.

For something to double in 10 years it only has to grow by 6.9% per year. This is from a somewhat unusual application of The Rule of 69. Latter is way of determining doubling time of something that grows by x%: just divide 69 by x and that will be the doubling time in periods.

So if the DB grows by 10% weekly, then it will double in 6.9 weeks. If the interest rate is 3% then the debt will double in years, etc.

The Rule of 69 is good for a approximation, that works well with low single digit increases.

Anyone who tells you to expect a 10% annual return from the S&P500 is lying to you. And probably trying to sell you something.

Really, I do not understand your reasoning since the S&P 500 has an annualized return of >10% since 1970.


Just because the S&P has seen such returns over the last 40 years, doesn't mean you'll see those returns over the next 40 years. The US economy is aging out, has stagnant wages, etc.

Also we saw a huge drop in interest rates that won't happen again over the next 40 years.

Going to disagree with you here. I think it'll be impossible to raise interest rates much further without causing economic damage. See negative interest rate policies in Japan and Europe, and close to zero rates still in the US.

I 100% agree with you about the difficulty of raising rates. But I don't think they'll continue to drop like they have over the past 30 years.

I have a hard time imagining a world of -5% to -10% rates!

> I have a hard time imagining a world of -5% to -10% rates!

A year ago I would've agreed with you, but with the ECB at -.04 and Japan at -0.02, I believe it more likely than previously.

Heh, fair enough. It's certainly very surprising to see rates below zero. So, indeed, I am much less certain about the future!

Incorrect, please re-read my statement. An average 10% annual return over 10 years. As the other reply to this notes, history shows in general this rule has held since 1970ish.

Don't forget about taxes. Interest on that 6% loan get's deducted from your income which is an arguable subsidy. Alternatively, you also avoid paying taxes on your S&P returns.

In many cases solar is really close to a no brainier. Which is why Florida Utility's have tried so hard to block residential adoption.

> In many cases solar is really close to a no brainier. Which is why Florida Utility's have tried so hard to block residential adoption.

This is actually turning into a hollow victory for Florida utilities. To bring people up to speed, Florida outlawed homeowner PPAs [1] (which locks out firms like Solar City, who install the panels and sell you power; no out of pocket cost for you, and you get solar immediately).

With the rapid decrease in panel costs, and the uptick in firms who are providing extremely cheap (or even zero percent loans in some programs!) financing for solar, Solar City (and other installers) are moving towards a financing model, which kills Florida's efforts to stop the spread of distributed rooftop solar.

Disclaimer: I did some volunteer work for the non-profit who attempted to fight utility legislation in this space [2].

[1] http://www.politifact.com/florida/statements/2015/jan/16/flo...

[2] http://www.flsolarchoice.org/

The accelerating payments in the PPAs might not be so great if electric prices drop due to solar (in the example below they go up 2.9% per year).


Being responsible for the full 20 year term is also a somewhat onerous financial agreement.

> Being responsible for the full 20 year term is also a somewhat onerous financial agreement.

I don't think so. Is it wrong to be responsible for a car when you lease it? Different asset, different terms.

A car is usually a much shorter lease with much better liquidity. There is a robust market for transferring a vehicle lease, the PPA is as likely to derail a sale of the house as anything else.

Usually, it's a more complicated issue, something like: "Historically, grid maintenance costs are amortized over electricity consumption, but when some people become net contributors, that breaks the model, since usage isn't high enough to pay for maintenance."

Of course, the correct solution isn't to compound one perverse incentive (out-of-sync fixed vs variable pricing) with another (ban on solar or refusal to buy); it's to change to a system of decoupled maintenance and power pricing.

The concern runs deeper than this. You can go off grid with solar for minimal additional costs, so they fear people will simply disconnect from the grid. Couple this with for profit electric company's and bribing/lobbying is the 'solution'.

this argument kind of annoys me.

"yeah, but how long before you see a return on investment?"

how about instantly?

this is what we did:

1) calculated our monthly electricity payment: $160 or so a month.

2) priced out solar to push our payment to net-zero-ish: the savings really start multiplying once you get down into the regulated tiers.

3) replaced our electricity bill with a solar installation loan.

what did we get in return?

instant insulation from rate hikes: http://business.time.com/2011/12/16/plugged-in-then-pay-up-e...

instant equity on our house.

oh, and look! i can now have a solar-powered car!

but yeah, if i happen to still live in this house in 15 years, sure, no electricity bill. and, i'm sure somewhere along the line i'll end up "making all my money back".

Re: Equity

Solar leases bring down the resale value of a home. If the equipment is owned it raises the resale value of a home.

I've sold 2 homes this year with solar leases. Both of them were great deals for the buyers netting basically the value of the lease off the full retail value of the home.


As solar prices go down, surely the added house equity goes down with it?

But what happens when you no longer get free power under net zero agreements (which is going to happen), and the cost of the system drops, reducing the value further of the already depreciating asset?

They will continue to save money on his monthly power bill, to a lesser extent. They will still have the warm fuzzy feeling that their existence is putting less of a drain on the planet's resources than they otherwise would have. And they may have captured a great deal of surplus value, possibly even enough to pay for the system before these changes are implemented.

In the past 25 years, electricity rates in CA have seen a 3-4x increase. If this continues, then even as some of the advantages drop away, the comparison function (monthly electric bill) will also increase, likely offsetting a lot of lost value.

> instant equity on our house.

Depends. Solar doesnt necessarily add value. Depends on the appraisal. I only know because I tried to buy a home with solar panels. The homeowner had raised the price of the home to include his investment in the panels but the appraiser wouldnt include the panels in the appraisal.

This is true, because the job of the appraiser is to estimate market value. If buyers will not pay a premium, they have no resale value. But in my experience buying a home in SFBA with solar, and getting it appraised for refi, it counted both times. (In the cost approach it was valued at $20k in 2013 and $10k this year)

Appraisals are not strictly tied to price of homes. Only the value for the loan. That's not to say you're not going to associate the two together (if you're buying a house with a loan), but the $30k one way or the other isn't a huge percentage of a 30 year loan.

Depends on where you buy. A 200K home in the midwest, 30K is a big percentage. The assumption being made here is that investing in solar automatically raises the value of the home. I am saying not necessarily. The example above was an appraisal in Madison WI, which, though in the midwest, definitely has plenty of solar installs.

Appraisals are supposed to be strictly tied to the price of homes. The appraisal shows how much the property would probably sell for on the open market today. The loan amount doesn't even show up on the report.

Payback of 6 years with the new cost, even for your low electric bill.

Excellent example of why solar panels (in the right locations) and LEDs are a better investment than the S&P500 right now. You spend dollars now to eliminate the need for future dollars.

While that is true, keep in mind that future dollars are less expensive than present dollars.

But it stands to reason that the prices of energy will rise faster than the rates of inflation, so it would still be a good investment.

If everybody buys solar, electricity prices will only rise faster than inflation in periods without much sun.

It might even be profitable to play a 'buy when it is free or cheaper, use when it costs money' strategy. You can get there by buying large batteries, charging them from your neighbor's solar panels when there is lots and lots of sun (they may even sell at prices below zero), and discharging them, selling to you neighbors, when there is little sun and wind.

I'm already investigating buying 50+ MW at a time of Tesla Powerpacks in certain markets exactly for this purpose, as an independent generator. Buy power when people are paying you to take it, releasing it back to the grid at the most expensive times.

Unlikely. Solar Panels don't replace all of your electricity costs. Naturally, you'll still run your refrigerator at night, right?

And during the day, when you have peak-solar energy, you'll be at work and won't be taking advantage of it. You'll recoup some of the costs through net-metering, but you usually buy back at wholesale prices and not at retail prices.

Finally, I'd expect utility companies to change their pricing scheme in the near future to account for solar. They service your electrical wires for example, and you'll likely have to pay a certain amount of money for that. So while you're not buying "power" (or at least, as much power) as before, you'll still have the service cost of being hooked up to the grid.

Chuck's sibling reply is spot on, but something else you must realize is that both renewables and energy storage are only going to get cheaper.

This means if utilities attempt to squeeze people with solar on their roof, they will simply move to batteries eventually. This already makes sense, today, in Hawaii (where the local utility was stalling on additional grid tie capacity, and with their per kwh rates so high, is was cheaper to go off grid). The more people who move off grid, the less people the utility can spread its costs across, increasing the rates for those remaining (which then rapidly incentives those people to move off grid).

This is called "the utility death spiral". Its a very real possible outcome, and utilities are aware of it. [+]

[+] http://www.greentechmedia.com/articles/read/this-is-what-the...


But what's to stop utilities from purchasing or creating cheaper large-scale projects?

I know you've seen my previous comments on Reddit with regards to Redox Flow batteries, CAES, and Pumped Hydro storage technologies. Utilities simply have access to battery-technology that the common people do NOT have access to. Some energy storage technologies only make sense in larger-scale 50MW applications or more.

If utilities serve a cooperative role in energy service, they will switch to the most cost-effective technology BEFORE the typical homeowner can switch.


Basically, solar pricing is moving FASTER than rooftop solar. Rooftop solar will always require insurance and roof-workers, while solar utility companies can just buy out a field in the middle of no where and install solar on the ground.

Cheaper and safer for utilities to work with. And then everyone in the town benefits, instead of just a few rich people who can afford panels (and the insurance of the rooftop workers)

At some point the price for the panels will drop below the price (and losses) of the transmission lines to transmit the energy from the large-scale project to your house. This happened long ago if you live in certain parts of the Rocky Mountains, and it will happen in areas of progressively greater population density.

Rooftop solar only needs roof-workers and their insurance until you have a cherry-picker-style robot arm to stick the panels up there with.

Utilities have a lot of capital tied in their existing fossil fuel plants, and their economics are probably tied into running these at capacity for their remaining lives.

So from their perspective it makes sense to lobby politicians to stall residential solar as long as possible, the environment be damned.

Last time I checked, it was the motivation of utility companies to make money.


Look man, you may not notice it. But the big utility companies out there are investing heavily into solar.

You need to consider the economics of "grid tie" you generate excess power into the grid and "buy it back" when you aren't generating in excess. So if you generate as much power as your going to use for the entire day from your panels, you pay nothing for that day. Even though you are drawing power at night.

In the Bay Area, there are other things that screw up the numbers but the program "trues up" once a year (rather than monthly) so some months (usually late spring through early fall) I'm generating more power than I use for the month, and fall/winter I consume more power than I generate in a month. The target is to hit it at exactly 0 which was easier to do when PG&E would exchange kilowatts for kilowatts but it is harder now with their system of charging extra kilowatts over baseline. Its screwed up and not fair but I get that they are hurting, if I could get Elon to sell me the battery pack out of a Model S and the inverter infrastructure to go with it, I could leave the grid entirely and that would be easier.

That's the idea with the Power Wall, https://www.teslamotors.com/powerwall Unless that's what you're pointing out

> You need to consider the economics of "grid tie" you generate excess power into the grid and "buy it back" when you aren't generating in excess.

Depends on the state.

Sometimes, you buy at retail prices and sell at wholesale prices. Which honestly, is far more fair. You're basically using the utility's batteries / energy storage at no charge right now (if you're in a state with pure net metering laws)

And yes, purchasing a PowerWall is more "fair". Instead of relying on the service and tax-incentives to give you a free battery pack, its definitely more fair if you got your own.

"Fair" in the context of a heavily regulated utility should have some relation to social benefit.

It's not fair for utilities to force costs onto consumers just because of a differential of political influence. It's not fair that solar reduces grid load at their peaks and saves them money on building out grid capacity and expensive and inefficient CO2 producing gas peaker plants and that benefit is not returned to the provider of that service.

Overall the many benefits of rooftop solar can be calculated and a value assigned. Most seem to come out at net or above, so talk of average wholesale prices is misdirection.

Consumers should pay a fair price for the service.

If consumers are using the utility's batteries and the utility's network, then the consumer should pay for the use of those services.

Peaker-plants can have a very simple solution: carbon taxes. Simple enough, and I support this measure to handle the externality.

Rooftop solar currently is probably a net benefit as long as the storage issue isn't a problem. (IE: the rest of the neighborhood doesn't have solar). But once solar adoption is widespread, someone needs to solve the storage issue (or "energy waste", if there is no more storage)

I'm glad you support a carbon price, because that's another great example of where "fair" becomes a political discussion.

I could say that if a coal producer wants to dump sulphur, radiation, C02 etc. into the air then it's "fair" that they recompense the people who live in the countries they pollute and the people they help to kill. Why should those who don't do this (people who generate their own electricity or pay to have non-coal energy) have to shoulder these costs?

Simple argument right? But no, it's a decades long political fight with no end in sight, because some groups have dispersed interests, while other groups have very direct incentives.

The same is true here, lots more rooftop solar would be of small benefit to most people. But it directly threatens some concentrated interests. And they will happily tell you that solar will destroy the grid, or net-metering isn't fair etc. etc. Not because it's true, but because it's in their narrow and short term interests to lie.

The difference is that I also think that net-metering is an unfair deal to utility companies. Again, someone needs to build the energy storage mechanism. You can't legislate away reality.

Unless it's dorm-fridge-sized, you can definitely run your refrigerator only during the day, generating ice to cool it during the night. Building HVAC systems often work this way, running the chiller to generate ice during the night (when electricity prices are low or even negative) and running the resulting chilled water through heat-exchanger coils during the day. And superinsulated fridges like Sunfrost models can stay cool during the night even without such extra thermal reservoirs.

A thing I haven't seen, but which ought to be practical, is a solar thermal fridge — for example, using the ammonia-absorption cycle that propane-powered fridges use, but heated with sunlight instead of flames. PV panels are typically 16% efficient, but it's easy for solar thermal collectors to reach 50% efficient, and they're also about five times cheaper per unit area (still). So you'd think it would make sense to power the fridge directly from sunlight.

As an EMH guy, I'd like to think that if an Einstein–Szilard cooler was more economical than grid power for A/C and/or food refrigeration that there'd already be systems for sale.

There are absorption refrigerators for sale for food refrigeration — I had one in my Volkswagen — but I don't think they use the Einstein-Szilard design, but a simpler one. Here's one: http://www.amazon.com/Norcold-Inc-Refrigerators-N841-Refrige...

> And during the day, when you have peak-solar energy, you'll be at work and won't be taking advantage of it.

I'm in Central Texas and I have two dogs at home. Even if I'm not there, I still leave the air condition going so they are comfortable. The last time I checked, my A/C system was the biggest power hog and so my electricity consumption aligns very closely with when the sun is shining.

Biggest power hog? Yes.

But not the only thing. My point is that its way too idealistic to assume that 100% of the costs will be covered.

I mean, yes, the laws as written seem to encourage a 1-to-1 transmission of energy. "Net Metering" is a law that subsidizes decentralized solar energy.

But in the future, when the subsidies run out and the laws are written to be FAIR (instead of written to encourage solar, as they are right now), you will not be allowed to purchase electricity and sell electricity back at the same prices.

All markets have a bid-ask spread. Wholesale prices are always going to be cheaper than retail prices.

I'm not sure what you mean by 100% of the costs will be covered. All I know is that I can reduce the amount of electricity that I buy during the day when I consume the most. The savings are greater than the cost so it's pretty much a no-brainer for me at this point. If I finance, I can pretty pay for the system with the money that I'm no longer using to buy electricity.

> when the subsidies run out

Why stop subsidizing solar? Do you think all energy subsidies are going away, or just solar?


I think that net-metering is going to go away once solar adoption reaches a certain critical mass. You can't keep giving away free batteries to consumers, someone eventually will need to pay for those batteries.

Other solar subsidies may remain a bit longer. But net-metering is probably the biggest once.

> I'm not sure what you mean by 100% of the costs will be covered. All I know is that I can reduce the amount of electricity that I buy during the day when I consume the most. The savings are greater than the cost so it's pretty much a no-brainer for me at this point. If I finance, I can pretty pay for the system with the money that I'm no longer using to buy electricity.

SunEdison's model has been proven to be incompetent. SolarCity's "MyPower" loans have been scuttled. That's WITH the current large number of pro-solar subsidies that the US Government is paying for.

I think that it's a "no brainer" to take advantage of SolarCity's deals as a consumer. But in the long-term, it looks like their finances are unsustainable.

Personally speaking, I'm more concerned about a long-term reliable model that manages to get the energy to the most number of people. Utility scale solar seems to be the best solution, although its a bit boring.

I'm not sure that the net metering argument applies to places like central Texas. People consume the most power when the sun is shining so solar, even without net metering, makes sense. Of course other regions are different.

> Utility scale solar seems to be the best solution

I don't think it makes sense to talk about a best solution. Large scale solar will contribute along with nuclear, coal, natural gas, wind, etc...

If you have space and money, you can create huge "gravity lamp" to store energy: lift up something massive using solar energy and electric motor, then it will slowly generate electricity at night using same motor as generator.

Or you can use "molecular spring", which is very compact, for same purpose.

How close are we to actually manufacturing molecular springs? My understanding is our materials science and engineering is not up for making macroscopic molecular springs yet, and what I can Google up seems to indicate we nowhere near macroscopic scale springs.

Make container of highly porous material then fill it with a fluid which will avoid that material surface at molecular level. AFAIK, it is invented years ago.

Can you post a link to that please? When I search around starting with the Wikipedia [1] entry, I don't find a description of what you are describing.

[1] https://en.wikipedia.org/wiki/Molecular_spring

> you'll be at work and won't be taking advantage of it.

Most people don't work in the dark

And presumably a lower bill implies lower energy usage, so he'd need fewer panels to cover it.

Ops point was that the original price of 40k in 2003 is now 7k for the total installation. That will change your math somewhat :) So even with your low power bill it's more like 6 years.

You can also sell electricity generated by your panels back to the grid.

Effectiveness of this varies by state or region. For instance, when I was in the Bay Area 10 years ago and looked into this, PG&E would let your electricity bill get to 0, but wouldn't let you go into rebate territory. They just wouldn't pay you for the additional electricity generated. I think other states/utilities have different options.

Most solar quotes take this into consideration, and quote you for enough solar to get into the minimum bill for your region (most utilities have a few hundred Kwh you can use per month at an obscenely low rate, and then it ramps up from there--solar is quoted so that you'll always stay within that cheapest range.)

For my current house in Austin, I'm still interested in solar, but it was both more cost-effective and had a shorter payback time to re-insulate to R36 in the attic and redo our horrible system (which had no ducts at all--agh, stupid 1956 house) than it was to do solar at the time.

Sadly, not in very many places. In most US states, for example, utilities have lobbied to get laws that don't allow residential customers to sell power back to them, at most net-metering agreements will allow you to get credits from any excess power your panels generate (which is how my utility handles mine - unfortunately, since I generate more than I use, I can't actually make use of those credits fully, and I'm giving away some power to the grid).

ROI can only be one factor. The other is that you are suddenly using clean energy. Doesn't this make you feel better?

The typical American median household income is $51k.

Spending $40,000 of that on "clean energy" so that they can "feel better" is well outside of the typical person's price range. Even $7000 would be cost-prohibitive, especially if they're paying or saving up for their kids.

A cold can of beer also makes you feel better. Its a hell of a lot cheaper too. The answer for the typical American is to wait for utility-scale to build Solar out for them.

When the city as a whole notices that solar is cheaper than running peaker plants or whatever, then the city / municipality can purchase an entire field of solar (AND the energy storage needed to make it useful) and give it out for the whole town.

And in the mean time, you can drink your beer on the porch while you wait for progress to be made.

US$40K was the cost when Chuck did it in 2003. Now we're talking about US$7000 over 25 years, which is US$23 a month. US$23 a month is not "well outside of the typical person's price range" in the US.

If you want to "save up for your kids", the way to do it is to invest your money in the investments that have the highest ROI, as long as you have enough liquidity to weather emergencies. Pay off your 23% credit card before your 17% credit card, pay off your 17% credit card before you start putting up solar panels that will have a 16% internal rate of return, and put up the solar panels before you buy the cold can of beer with its 0% ROI.

Twenty years from now, you'll be drinking your beer on the porch and sending your kids to college, while your typical American loser neighbors are whining about their credit card and electric bills.

> US$40K was the cost when Chuck did it in 2003. Now we're talking about US$7000 over 25 years, which is US$23 a month.

Chuck is replacing a $2000 inverter in just 10 years. If your emergencies include "fixing the damn solar panels that you bought", that's your own damn fault.

In any case, when middle-class Americans start buying solar en masse, I'll admit I'm wrong. But for now, I'm going to bet you that this is a luxury that is not cost-effective for the typical middle-class American.

It ain't like a Washing Machine, Refrigerator or Water Heater. I'll tell you that much.

It ain't like a washing machine in 2016, which everyone already has one of. It ai very much like a washing machine in 1916, when not many people had one yet, in that the year-over-year growth rate in installed solar panels is at about 26% (http://www.economist.com/news/business/21696941-solar-power-...).

An inverter that cost $2000 ten years ago now costs $400. (Amazon has a Bestek 2000W inverter for US$140 right now; three of these can supply more than what my house is wired for.) Even if you did have to buy a new $2000 inverter every 10 years, that's US$17 a month.

It's true that financing the up-front investment is a difficulty, and that's why Solar City is doing such great business. But as the component cost drops, accelerating both the IRR and the up-front investment size, that's less and less of a problem.

If you do want to bet on it, I imagine you can buy stock or long-expiry call options in coal-mining companies (BTU is in penny-stock territory right now because they filed for bankruptcy a couple weeks ago, and ACIIQ is too because they filed for bankruptcy in January), or puts on SCTY and Enel Green Power. This is a terrible idea and will lose you money, and you should not do it.

Because you refuse to calculate, you are doomed to talk nonsense.

> It's true that financing the up-front investment is a difficulty, and that's why Solar City is doing such great business

On the contrary. SolarCity hasn't made a profit and their stock has tanked. It seems like solving the up-front cost issue is difficult.

Widespread deployment of Utility-scale Solar will harm SolarCity and their customers (who have entered into long-term purchase agreements). If Widespread Solar LOWERS prices in the long term, then SolarCity dies.

Good for the world of course, but SolarCity's bets are on the wrong side of progress IMO.

The up-front financing business is hard.

> In any case, when middle-class Americans start buying solar en masse, I'll admit I'm wrong.

Promise? Because that day will be here sooner than you think.


> The typical American median household income is $51k.

The relevant number is "Amercian median house-owning household income"

Solar panels don't have to be replaced annually. They have 30 year warranties as standard, and are likely to last at least 40 years. Spending $100 per month to eliminate your $110 (the us average) and rising electric bill makes sense even if you don't value renewable energy on principle. This isn't even considering the fact that you can sell renewable energy credits for the energy you sell back to the grid.


Why would electric bills be rising if solar is getting cheaper? If solar is cheaper and more efficient today, and utility-scale solar can afford efficiencies like motors (to directly face the sun throughout the day), and building the panels on the ground (easier to maintain, fewer rooftop deaths)... why would I want to jump the gun and purchase solar for myself?

Tell me, do you think you can beat $.06/kWh that Utility-scale Solar is getting TODAY? Do you think you can beat the estimated cost of $0.03/kWh in 5 years?

I THINK the article is trying to say that electric prices in America is going to start going down, thanks to solar. Depending on how quickly solar gets deployed of course.


Now yes, I do realize that buying your own solar comes with its own rewards. But we cannot ignore the risks. Gas prices have halved from their height of $4/gallon, and other forms of energy have also dropped in price.

Most noticeably, solar. No reason to buy solar yourself if you can just wait for the utilities to deploy solar after all.

First off, that 40K was the cost 10+ years ago, not today. Even so, lets be honest - the population of HN is not the median household. Even if the average American cannot afford to pay cash for solar, many software engineers probably can.

I've been trying to do this calculation myself. One factor I hadn't initially put in was the fact that I didn't need to think about a backup generator for our (east coast) storms.

You might—I do. Your inverter won't power the house when the grid supply is down. Why? Well, because if the grid supply's down, it's probably because there's a line down somewhere. Therefore there's a lineman trying to pick that line up somewhere. He's going to be handling both ends of a broken, downed line. The generator-side line is tagged out, that's safe. The house-side line? Well, that's up to you—and so your equipment will deactivate to avoid creating a safety hazard for grid maintenance staff.

An auxiliary grid independent outlet is apparently becoming a feature on newer inverters:


A generator probably beats a lot of battery systems (which along with the right inverter and a transfer switch is another way to use the solar for backup power).

Well one issue with grid-tied systems is that when the grid is offline they provide no power (they explicitly don't want to back feed into the power lines and kill some lineman who things power is off). So one of the certification issues is to make sure that your inverters recognize that state and shut down.

But a side benefit of putting in the system was we did a lot of electrical panel work, and as as result put 8 circuits on to a transfer switch (the incremental cost was the transfer switch and the electrictian's time wasn't materially affected.) So when the power does go out we can fire up a generator and switch over to gen power easily.

I keep a close look at the economic of rooftop pricing in TX - it's still not profitable switching but getting close.

Cons: - Retail cost of electricity is already v low in most of TX (Houston, Dallas) - so it's hard to compete. - not much competition in the solar rooftop installation space (due to first point above) - No net metering policies in Houston, Dallas.

Pros - High solar yield

Keep in mind that the 30% tax credit will eventually taper off. It might behoove you to wait now, but not forever.

What is this 30% tax credit you speak of?

Y but now it's 7K and not 40K.

33.3333 * 7 / 40 = 5.83 years

$100 a month in the summer in Texas? You living in a shoe box in the panhandle or what? I spend $700 a month in August. Tell us your secret for real.

I live in TX and have never spent anywhere near $700/m on electric. Either you keep your AC at 65 or something is very wrong.

And I'm sitting here in Germany with my €30/mo bill. :)

I live in a brand new home. Spray foam insulation. Keep the AC at around 74 year round. I also have natural gas running in to my house for my stove and water heater. All of my light bulbs are LED.

What are you setting your AC to?

> How long until you start seeing a return on investment?

Why can't the utility companies do the financing (upfront investment), place them on our roofs, and get a little bit of the profit? Seems like a win-win situation.

And otherwise government, or specialized banks?

Hey Chuck, I've been waiting for solar to come down and it sounds like it has. When you priced it, was it just for parts or was that $7K installed?

And for the do-it-yourselfers, how much of the cost is installation now? Is it practical to do most of the grunt work yourself and then get an electrician in to do the final part?

As mentioned elsewhere it was just parts. Installation of grid tie systems isn't hard but does have a number of building code things you have to follow (at least in California). At the time we did the install my wife took a solar installer's course which went over the current code.

I can recommend that as a plan of action, find a local group offering job training as a solar installer and take it. Then using that knowledge you can design an installation that meets your local code.

The huge power companies still control the grid in many countries though. I hope improvements in batteries/storage will remedy this and we'll shift towards community grids of sorts. You'll still need the traditional grid for peaks but the power is shifting (pun non intended). It's a great development, I haven't looked into the net power needed to produce solar/photovoltaics panels but it was dropping quite a bit as well.

I'm a bit surprised there are no mini wind turbines. I read about some promising developments (with close to 0 noise output) but it seems to have fizzled.

Would you be able to share what company quoted you for $7k? I'm going to be getting quotes on a solar installation, and your quoted price seems rather good. Do you get to reuse parts of your old installation?

Well in my case we already have the mounting equipment on the roof and wiring. And we were looking at new panels and a new grid-tie inverter, so 16 x 325W Kyocera panels was $5,600 and the SMA Sunnyboy 5Kw inverter was $1900 (total $7,500).

I would not expect a company to quote a new install for that price since they will have to add the mounting system and pay their installers for a day's labor getting everything set up.

[1] http://www.solar-electric.com/

Not too long ago some of the comments here were saying "how will we feed all of those millions of EVs out there if the EVs do end up replacing most ice cars?"

Well that's the utilities opportunity to survive. Utilities should actually lobby against ICE cars and pro-EVs. They will be the ones charging EVs everywhere. Not everyone will afford to charge their own EV at home where they have solar panels.

Other than the fact that $7k still isn't that cheap (for many people)--how long does the "average" home of 4 take to recoup that cost?

I think the barrier is not only cost, but also know-how. Few people are likely to tackle a challenging project like this with a $7k out-lay.

Why not replace your current panels with new ones? If you can generate nearly 3 times as much electricity with the same roof footprint and be able to reuse your racks and electrical installation, you should be able to be in the black in no time.

Sort of, and sort of not.

I am already "in the black" as it were since the system has already provided more value than it cost to install and maintain. So no worries there. But 325W panels are physically larger than 185W panels (not twice as large but still) so replacing the 28 panels on the roof with them would get my raw input up to 9.1kW but I'd need roughly 33% more surface area. That would add another strut on the mounting system and the top row of panels would extend beyond the peak of the roof (now disallowed by the building code). However, it would be "easy" get get 16 in there, which would be an equivalent raw foot print (5,200 watts)


Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact