The Los Angeles Department of Water and Power (LADWP) Board of Commissioners was presented with the Eland Solar & Storage Center in Kern County, California, from an LADWP internal team on June 18, 2019.
The team told the commissioners that on July 23, they plan to seek approval of a two phase 25-year power purchase agreement (PPA) priced at 1.997¢/kWh for 400 MWac of solar electricity delivered at time of generation plus a adder 1.3¢/kWh for the excess electricity later delivered from a co-located 400 MW / 800 MWh energy storage system, which would mean an aggregate price of 3.297¢/kWh “along with all associated Environmental Attributes, Capacity Rights and BESS Products but excluding Excess Energy and its associated Environmental Attributes, Capacity Rights and BESS Products”.
Per an email from 8minute, the project will be built in two 200 MWac solar phases. There is no price escalator, and the solar portion is a record low price for the United States. It even beats out the current U.S. pricing leader – 8minute’s 2.375¢/kWh from the 300 MW Eagle Shadow Mountain solar project.
The project includes the option to add 50 MW / 200 MWh of energy storage for an additional adder of 0.665¢/kWh. It was suggested that the excess electricity will be used during the evening peak period to ease ramping, though one presenter also brought up that the morning peak was as important to consider (see analysis of this volume at end of article).
The Kern County website holds thousands of pages of development documents – mostly related to environmental analysis – but some describing the installation itself. The project will be built among more than 1 GWac of solar power projects, including others developed by 8minute. The site documents state that there is more than 500 MWac of capacity available within various pieces of land in the Eland development proposal, meaning this site will probably expand with more solar as more PPAs are signed in the future.
The current record solar power price in the world was signed in Mexico at 1.97¢/kWh as part of a batch of projects averaging just over 2¢/kWh. And while there was a lower bid submitted in Saudi Arabia at 1.79¢/kWh, it wasn’t signed.
The LADWP representative noted that there were probably about seven other projects on their short list that they were in advanced negotiations with.
The project is expected to receive Note to Proceed with Construction in later 2022, with first installed capacity able to deliver in April of 2023, and a guaranteed commercial operation date of the last day of 2023. The presenting team noted that the project would represent approximately 5% of the LADWP RPS goal (below image) of 100% clean electricity by 2050.
Using Helioscope (pdf), and data from the Kern County documents, pv magazine USA did a rough simulation of a single 2.0 MW section, with 2.65 MW of modules attached using single axis trackers located at the site. The purpose of doing this was to determine the volume of electricity that could be captured from clipping if a 1 MW / 4 MWh battery was put behind the above layout.
This author does believe that the output will be even greater than is projected here as standard solar modules are modeled, while we expect bifacial on this project.
Note in the image below from the simulation – there will be approximately 2.7% clipping loss over the course of the year equal to ~181 MWh/year (called “Constrained DC Output” in below left chart). A 1 MW / 4 MWh battery that cycles once a day for a year can move 1.4 GWh/year.
Since these units will only be able to cycle once per day with excess solar power, they could also be performing other tasks in periods outside of the charging to discharging period. The presenters noted that the morning electricity ramp needs support as well, so there could be other services that the project delivers to LADWP.
Edits: This article was corrected and updated around 11:40 AM Eastern Time (U.S.) on June 28. F8minute has communicated there is no price escalator, and that the second phase is a full second 200 MWac / 265 MWdc + (maybe) 100 MW / 400 MWh.
This content is protected by copyright and may not be reused. If you want to cooperate with us and would like to reuse some of our content, please contact: editors@pv-magazine.com.
Wouldn’t the battery power be an additional 1.3 cents/kWh, so ~3.3 cents total? Otherwise the stored energy is selling for less than the directly generated solar — that sounds odd.
Separate product, so not added on top of it
its not stored energy my man it’s ” 1.997¢/kWh for 400 MWac / 530 MWdc of solar electricity delivered at time of generation plus 1.3¢/kWh for the excess electricity later delivered from a co-located 100 MW / 200 MWh (and maybe double’d in the second phase) energy storage system.”
once u put that bad juice in the grid it’s gone baby! simple physics my friend….
“here as standard solar modules or modeled, while we are expect by facial on this project.”
I assume this should be
here as standard solar modules *are* modeled, while we are expect *bifacial* on this project.
Ty
What are the other costs that are not included in this price? How do they guarantee a battery capacity for 25 years at a fixed rate? Do they just guarantee the capacity and replace on their end as needed?
I see bankruptcy in their future!! There is not a chance the battery will cost that little /kwh .
Or the solar for that matter, much less make a profit or even paying interests on the loan. At those prices interest rates are critical.
Unless there are massive subsidies I don’t know about.
There are massive subsidies you don’t know about. Be they nuclear, wind, solar, or good ol’ coal, grid power suppliers get huge subsidies to help them get up and running.
It’s not here yet! It’s a glance at what they hope to be selling power for in 2023.
Doing the math the battery will have to do 18k cycles not including interests, other costs to just pay itself off at Tesla Powerwall price, the lowest one I know of.
With all the costs would double that.
At 1/day that is about 50 yrs.
LADWP has a much larger, cheaper, and long-lasting battery: Pyramid Lake. They use excess solar and wind energy to pump water from Castaic Lake to Pyramid Lake during the day, and at night, it flows back, generating electricity.
There are other battery technologies for grid storage that will far outperform Li-ion. Tesla is pushing hard to make their batteries cheap because they need that for mobile use in cars, and so it’s a decent option for home backup. But grid storage doesn’t benefit from lightweight batteries, only from cost vs. storage capacity and longevity.
Many grid level projects are using flow batteries. The flow battery decoupled output capacity from storage capacity, cycling liquid electrolyte through cell stacks to charge or discharge. Need more output? Add more stacks. Need more capacity? Add more tanks of electrolyte. It’s also safer than Li-ion at those densities.
Well considering that Li-Ion can make about 1,500 cycles (optimistically) something here smells fishy. Also the fact that at something like 24% capacity factor and $540/kW (the cost of just panels), 90% DC-AC efficiency you are looking at something like 1.1 cents/kWh baseline. Realistically cost of land, inverters, installation, human monitoring, transmission lines, etc you are likely looking at 3-4 cents per kWh.
The EOS Aurora battery is the cheapest I know of.
$95/KWh and 5,000 cycles lifetime.
https://www.windpowerengineering.com/electrical/power-storage/eos-energy-storage-taking-orders-95kwh-eos-aurora-dc-battery-system/
But that is still $0.019/KWh just for capital cost of the battery. The prices in the proposal are crazy low.
Doable as battery system shaves the peaks and allows better solar production with less investment to power electronics. 1c/kWh is so-last-season after few years. 2030 it is probably already 0,1c/kWh.
John (above)
I believe that “bifacial” refers to the modules being capable of being tilted to follow the suns seasonal altitude changes, summer – winter, as well as rotating to follow the daily path of the sun, east to west. It makes a considerable difference in the panels output.
I misspelled a few things originally which is why the question about the weird version of by facial was brought up, I’ve corrected my misspellings since
Bi-facial need not be installed on a tracking system to get more daily “harvest” from the sun. IF you install a typical solar PV panel on the roof of a house, the sunlight falling on the surface of the panel creates the D.C. that will be inverted into usable electricity for the home. Bi-facial panels are usually transparent and any ‘side’ or reflected light from off of the roof also goes into the generation of power. It has been proffered that if one puts bifacial on their roofs, use light reflecting tiles, shingles or paint the roof white with polymer roofing paint for the best results.
Has there been any comment from 8Minute or LADWP?
There’s nothing on 8Minute’s News section of their website… https://www.8minute.com/news/
Are we 100% sure that this is real? Its a very big milestone to be buried on a few pages in a thousand page bureaucratic submission. Need to be careful to be 100% sure with market moving statements like this.