California Solar Duck Curve Gets Bigger | Hacker News Reader

California Solar Duck Curve Gets Bigger

cleantechnica.com

One possible solution for this is to encourage employers to install more charging stations at their office -- which would mean a more consistent load throughout the day as tons of electric cars charge during the day. This would also be a huge benefit to employees who could effectively drive their electric car for free and the power company also gets the benefit of people putting less demand on the residential power grid.

At this point my (weak) personal observation is that there are more EV charging spots at most parking lots (in SF at least) than there are EVs using them.

I think we just need to accelerate the rollout of EVs as quickly as possible.

Is this not basically equivalent to installing more batteries on the grid assuming the number of electric cars is mostly unaffected by the availability of employer-provided charging[0]? Those cars are batteries, and midday charging smooths demand the same as grid-connected battery storage[1].

[0] A big assumption, to be sure.

[1] You could schedule your car to charge at the most off-peak hours that aren't midday, but we're getting pretty far from a first-order approximation if we're looking at that as a factor.

Solar covered parking with charging.

If they install more vertical solar, it will increase renewable output in the mornings and evenings, flattening the duck curve.

Vertical solar can be deployed anywhere, works well with farms, consuming zero land.

https://sunzaun.com/flattening-the-duck-curve-with-vertical-...

https://interestingengineering.com/innovation/pioneering-col...

Someone should run numbers to see if this will be economically viable, since output during morning and evening will be much lower, meaning electricity cost will be much more expensive after offsetting cost of panels, installation and maintenance.

Output at the morning and evening should be higher with steeper panels as the sun hasn’t moved up as much. It is a hit to total production, but kWh aren’t fungible throughout the day.

The morning and evening kWh are becoming worth more than the mid-day ones.

To do the comparison, it should be compared roughly with the cost of building (batteries + traditional solar). I say "roughly" because batteries can do more than levelize solar, they can levelize other intermittent sources as well, and can provide power on foggy mornings and/or evenings, while vertical solar cannot.

Indeed in an off-grid build, I know someone that optimized their panels to produce earlier in the morning instead of maximizing production, because switching off battery power a little earlier was most valuable to them. Losing some throughout the day and evening wasn’t as much of a concern because they were primarily charging for night-time.

Focussing rooftop installs that have a more eastern or western angle, or steeper slope, could be a way of flattening this curve.

This only makes the "problem" of the duck curve worse, by widening the part of the day that solar displaces all other generation, and making the switchover point in the evening even steeper.

But that's fine, any "problem" potentially caused by the duck curve has long since been addressed by plumeting cost of solar PV and batteries and a host of other improvements put into place since the potential issue was identified.

The real problem of the duck curve is that it's become a misunderstood meme that fools people into thinking "we've built too much solar, too fast, which will cause vague problems" when that is far from the case and the number one thing we need to do right now and for the forseeable future to improve the energy grid is build more solar PV.

>The real problem of the duck curve is that it's become a misunderstood meme that fools people into thinking "we've built too much solar, too fast, which will cause vague problems" when that is far from the case and the number one thing we need to do right now and for the forseeable future to improve the energy grid is build more solar PV.

Absolutely! Deploy solar, demand will shift.

If new solar deployment is vertical, it will do two things: (1) duck curve won't get deeper (2) duck curve gets wider.

I wonder if the economics are distorted here. It looks like there is 0 demand during the day, but people are obviously using electricity during the day. Should we not make electricity actually free at the hours when there is too much of it, to encourage its use? This seems like the best way to encourage battery installation and electric car usage, without requiring every house to get solar panels.

The downside, I suppose, is that electric utilities are for-profit companies whose shareholders need a return. Thus it's hard to property subsidize. The power lines to your house cost money to maintain regardless of how much electricity is flowing through them. (Something I hate in general is this kind of usage-based billing that subsidizes fixed costs. You should pay a fixed fee to be on the grid, and then per-kWh for usage. But people hate being billed for things they don't use, so we try to combine to costs of the infrastructure with the cost of the incremental usage. It's really just a crazy security that should be in the domain of banks and not ordinary consumers, but it's also so normal and widespread we think nothing of it.)

Yes, grid fees should be a flat payment based on your max power draw in the last year. Or maybe a time dependent fee that reflects current power grid congestion.

Certainly not something that simply scales with the energy consumed.

Most utilities, including I assume California's, do have a flat component of the total bill to cover operational expenses like salaries, the fixed cost of infrastructure, hedging against huge spot price spikes, etc. In my case (Pacific Northwest), this amounts to about one-sixth of my total bill. But having 100% of your bill be a flat fee would just be crazy: there'd be no incentive to conserve energy and "based on max power draw" would encourage people to use 100% of their possible consumption all the time.

You misunderstood me.

The GRID FEES should be flat. They are just one component of the price that is charged by the utility, and currently they are charged per kWh consumed.

The POWER PRICE should of course scale with the kWh that are being consumed - either as a fixed price, or better with a component that depends on the time of the day or the actual market price.

Part of the grid is scaling with power usage, so I think it makes sense that someone who uses more power has to pay more for grid upgrades/maintenance.

Grid is scaling with max power draw at times of congestion.

Congestion is heavily correlated across consumers, so you cannot prevent it by charging higher marginal power costs.

(Like mobile carriers do with data)

Isn't that the same idea as charging more at certain times of the day? We can bake grid fees into those peak hours.

Then if there is equal usage between all consumers like you say, then everyone would get charged equally for the grid . Functionally the same as a fixed price.

FWIW I'm in California and my bill is totally based on usage (although not linearly). However in a few years the state is moving towards adding an income-based fee

I'm on SCE in California and I pay about $0.73 per day in "Basic Charges" (about $23/mo) if I use zero power. There are also grid-related charges (including a wildfire fund charge) that are usage-based on the order of $0.03/kWh. For my usage (about 1MWh per month) that means half the grid charges are flat and half are from usage.

For comparison, my TOU energy rates range from $0.06 to $0.25 per kWh depending on the season and time of day.

Larger consumers usually do pay demand charges.

In the UK, they had some method to charge based on a retrospective look of when the grid had the largest demand and your contribution to it, so some firms were set up to help you predict this and reduce your usage at those points.

> The UK’s demand charge price is based on its ‘Triad’ system. ‘Triads’ are three, half-hour periods in the UK winter when energy demand is at its highest.

> The Triad dates must be at least ten days apart. They typically fall on weekdays, between 4 pm-7 pm, November to February and during periods of cold weather. The National Grid often has to fire up expensive coal and gas-powered stations to meet industrial and domestic demand during these windows.

https://www.greenoptimistic.com/how-to-reduce-commercial-ele...

This all makes sense for larger users but the costs of smartmetering on light users has been a bit of a farce in Ontario Canada where the costs were real and the load shifting was not.

The Duck Curve is from the point of view of the utility so it not only removes rooftop solar, it goes further and removes utility owned renewables, to focus on the problem of a) turning off fossil production at points during the day and b) dealing with the ramp in fossil demand in the evening.

> I wonder if the economics are distorted here. It looks like there is 0 demand during the day, but people are obviously using electricity during the day. Should we not make electricity actually free at the hours when there is too much of it, to encourage its use?

In my state (Queensland, Australia) demand is actually negative during peak production hours on many days.

The wholesale price actually goes negative during these times - that is, if you are a wholesale consumer, you can get paid for using energy.

Individual households can opt to go on to wholesale plans - my uncle does it (in Sydney, much further south) as he has way too much solar and battery capacity for his household to use. Been a whole since I spoke to him but I know he had a couple days where he made $20. (He is a retired nerd and spends a lot of time optimising usage patterns to maximise money.)

This kind of thing will become way more common, especially as automation gets better and storage becomes more prevalent.

I’ve always wondered why the price of electricity on a time-of-use plan doesn’t match this curve. Our electricity is still “peak” pricing in middle of day at the bottom of the duck curve.

The duck curve chart is a bit misleading - it is "net load (or the demand remaining after subtracting variable renewable generation)", not total demand.

Total demand is still highest mid-day, but there's enough solar being generated to lighten the net load.

But your point is basically apt: As more and more at-home solar comes online, grid-demand WILL drop during the day and pricing SHOULD drop accordingly (but we all know it won't unless the utilities get a legislative requirement to do so).

Duck Curves that you see published are generally from the spring, (march to may in this case) when solar production is high and air-con demand is low. Time of Use rates are generally set anually, so need to account for the summer season.

It's an artifact of previous usage curves when PV was negligible.

My utility has new peak hours starting around 6. I'm grandfathered in on the old plan for now, which is a difference of about $300/mo. in the summer.

It is? Poking around it looks like peak rates are typically set from 5-9PM (https://www.pge.com/en/account/rate-plans/find-your-best-rat... for example).

Fortunately, they are adding massive amounts of battery capacity to compensate.

Gridstatus.io is a great place to watch it in action: https://www.gridstatus.io/home

Right now batteries are 10% on CAISO and natural gas is 42%. That's closer than I thought it would be, but it's just after sundown. Later tonight it will drop down to 0%. This suggests it need 5x the power capacity and significantly more than 5x of the energy capacity it has today.

IIRC, the battery discharge increases again in the early morning hours to compensate for the morning load peak.

That's exactly what I'd expect to see, if the goal is to replace peaker plants.

Why are they not using the excess solar power to lift large weights above the ground that can be released and lowered to power generators during the night?

Setting up large enough weights to lift is incredibly expensive - it’s a terrible way to use power, let alone store it.

Just make them out of water.

Still very expensive you don't have geography that gives you elevated storage containers for free.

Dig a hole!

Because the physical infrastructure to do that is very large and expensive.

To put it in perspective: a cubic yard of concrete costs roughly US$125, and the amount of energy you can store by lifting one of those cubic yards 10 meters into the air is worth something on the order of $0.01, assuming perfect efficiency.

I pay about 20 cents per cubic yard of water at home.

Which is pretty fantastic, all things considered. You pay 20 cents for 1,700 kilograms of safe, drinkable water, which is enough to comfortably last 500 days for one person.

They have cheaper options to use first like:

* exporting to neighbouring grids

* charging batteries

* reducing hydro output

Which you can see in action here today:

https://www.caiso.com/TodaysOutlook/Pages/supply.html#sectio...

(You can change the date to March-May to see it actually go negative on imports, i.e. export excess.)

California has pumped hydro, so they are.

This is done in some places in the world, water gets pumped into reservoirs that provide energy again when they discharge.

This is being experimented with. So far it seems to work best in abandoned wells which provide a controlled space for horizontal movements.

Good to see solar power getting more common, but we need batteries and storage to truly end carbon emissions.

Yes. And the consumer is meant to buy the batteries. And these (expensive) batteries only last 7 years of so.

You could say, electricity production/generation is being outsourced to the consumer, who will still pay bills, but now will also assume some of the infrastructure costs too.

Where the heck did you get that idea from? The batteries discussed in this article are storage stations that are part of the grid, not something bought and maintained by the consumer.

because NEM 3 makes new rooftop solar installations uneconomical without batteries. though even with batteries they're still pretty uneconomical

Because NEM2 wasn't sustainable long term. TFA is talking about we don't have enough batteries for all the solar and how it might be necessary to reduce solar production during peak times. That's why we don't want to keep incentivizing solar without batteries.

I'm responding to a comment that mentions batteries and talking about personal experience, for a home solution... This is relevant as a lot of solar uses private roofs.

In the UK, you could sell all your unused excess roof solar energy back to the grid. But this is at a wildly reduced rate - fractions of the cost you would pay for electricity. So you would then need batteries to store the energy... But then you incur the drawbacks of these (high maintaince costs).

With all these personal solar solutions, with batteries, panels etc being bought and provided by the consumer, it seems an obvious point to me that quite a lot of electric infrastructure is moving to the private domain, with discounted electricity then being sold back to the public utilities.

Solar+batteries remains insanely expensive putting nuclear to shame (nuclear in many areas is still way cheaper than solar without batteries)

Some even say the duck curve looks more like a "canyon curve" now.

https://www.powermag.com/epri-head-duck-curve-now-looks-like...

relevant:

https://www.tesla.com/utilities

Reply on news.ycombinator.com