#1 cleantech news, reviews, & analysis site in the world. Subscribe today. The future is now.

Clean Power

Published on March 18th, 2016 | by Susan Kraemer


CPUC Gets It Right: PG&E Can Keep Its Ivanpah Contract

March 18th, 2016 by  

A completely innovative technology that is one of the keys to slowing climate change was today allowed by the CPUC to fine-tune energy production. The decision results in no harm to ratepayers.


The California Public Utilities Commission (CPUC) has approved PG&E’s December 18 request for a Forbearance Agreement for Ivanpah Units 1 and 3, giving the two units at least six months, and possibly a year, to meet current production targets of 448,000 MWh annually.

In unanimously approving the resolution, the CPUC noted that the plant has “substantially increased” its production.

The production targets escalate for a four-year ramp-up period till 2018, when 640,000 MWh should be generated by the two units under contract with PG&E. But there is a ramp-up period of 4 years.

Under averaged annual generation over its first two years, according to filing documents quoted by KQED’s Pete Danko, Units 1 and 3 were contracted to generate 70% of the 640,000 MWh; or 448,000 MWh.

Ivanpah I

Ivanpah III
In the first two years, according to the EIA data seen above, Ivanpah 1 and 3 generated an average between both years of 361,576.5 MWh, falling nearly 20% short of the targeted 448,000 MWh. But 2015 ramped up considerably.

Adding up Ivanpah’s — much improved — 2015 output, generation increased to 433,331 MWh of its 448,000 MWh target; around 97% of the goal.

“There was a more than 50% increase in production during the second year of operation compared to the first,” said Joe Desmond, spokesman for BrightSource, the initial developer and solar technology provider of Ivanpah, now operated by its partner NRG on behalf of Solar Partners (Google was the third partner).

Desmond told Cleantechnica that the Solar Partners are confident the plant can ramp up to full production by 2018 as originally planned.

“Now entering its third year, this first-of-its-kind solar thermal project continues to set new production records,” he said.

In December, PG&E had petitioned the CPUC to approve a Forbearance Agreement between PG&E and Solar Partners, giving more time for Ivanpah Unit #1 and Unit #3 to reach the contracted percentage of targeted generation under the PPA.

PG&E’s December filing stated:

“The Forbearance Agreements benefit PG&E’s customers because they:

(1) ensure the continued operation of innovative solar thermal resources that provide Renewable Portfolio Standard (“RPS”)-eligible energy;
(2) require the Solar Partners to pay consideration, which is not currently required under the PPAs;
(3) allow the Projects time to continue to improve performance to meet the requirements of their respective PPAs.”

PG&E’s contract had envisioned a four-year ramp-up, because of the size and degree of innovation in Ivanpah.

As the largest CSP tower in the world, Ivanpah had over six times the capacity of its only US predecessor at the time. Nevada Solar One, a single tower, was just 60 MW net. (It also used molten salt-based technology, unlike Ivanpah).

Ivanpah was an enormous jump, as the first to use purely water-based thermal power towers, and the world’s largest CSP tower project, going right to 390 MW gross, of which 377 MW is available for net generation to the grid.

What “Parasitic Power” Means

All turbine-driven power plants — including gas and coal, hydro, geothermal and CSP — normally include some extra generation to supply “parasitic” (onsite) needs, such as for condenser cooling or overnight power to keep turbines turning slowly overnight so they don’t crack upon morning startup or (in CSP) moving the heliostats.

Ivanpah also added up to 5% natural gas to supply some of this overnight onsite need. (It had no storage to take care of parasitic power, as PG&E had no interest in storage back in 2008, thinking pumped hydro at Hetch Hetchy was plenty back then.)

If fossil energy gets ramp-up periods, why not innovative clean energy tech?

It is erroneously suggested in media accounts that Ivanpah’s ramp-up period is unique, and that Ivanpah asked for it after the fact.

“In fact, a multi-year ramp-up period was always assumed in the financial and operating plan for each unit,” Desmond explained.


BrightSource’s 2010 DOE filings had already noted that “initial performance will be less than full design,” then would rise due to “realization of the operator’s learning curve, procedural optimization, and fine-turning of equipment and systems for increased plant performance.”

“The DOE was made aware of the ramp-up period in connection with the Loan Guarantee application, as were the project’s equity investors and PG&E who negotiated the terms of the PPA,” he explained. “Moreover, the California Public Utility Commission approved the PPAs which included those target dates.”

Ramp-up periods are actually routine in traditional thermal carbon-combustion technology (where instead of leveraging simply sunlight and water, carbon combustion “burns up the fenceposts” as Edison put it). What is burned in such cases is gas or coal, of course.

“Even conventional systems, using tried-and-true technology, are afforded 180 days after commercial on-line dates before EPA presumes they are in full operation,” Desmond pointed out. “And that is just the default; they may petition for additional time if they have not yet reached their potential.”

(Kevin Smith, CEO of SolarReserve, who previous to developing Crescent Dunes CSP, had spent 20 years in the traditional thermal energy business, confirmed that traditional thermal plants normally also are expected to have ramp-up periods.)

PG&E Battled Opposition to Keep its Ivanpah PPA

In light of the demonstrated improvement as reflected in the EIA numbers, PG&E had asked the CPUC to approve a new forbearance period to allow Ivanpah time to meet the contracted generation percentage for this pre-2018 generation.

The CPUC is to be congratulated in making the right decision, in consideration of Ivanpah being so close to the first two-year target by 2015, given the data on improvements in performance. The decision was unanimous.

The Office of the Ratepayers Advocates (ORA) previously fought net metering for rooftop solar, contending that grid maintenance costs are increasingly being dumped on non-solar customers.

ORA attempted to get the CPUC to refuse PG&E’s request and to force it to simply terminate the PPA as too expensive. In response to ORA objections, the PG&E stated in Reply to Protests of AL 4761-E.pdf:

“The Forbearance Agreements avoid the time, cost and uncertainty of dispute resolution by providing a limited period during which the Solar Partners can demonstrate that the Projects will be able to meet the Guaranteed Energy Production (‘GEP’) going forward.”

CSP is more expensive than today’s PV. But PV was nearly as expensive in 2008, and ORA has not objected to those old utility-scale PV PPAs at those much higher rates than today.

Like PV, CSP has lower costs in more recent plants, within the global markets that kept awarding PPAs after 2008 (when the US stopped awarding CSP). The most recent CSP prices bid in global markets are now under 10 cents per kWh — and these include 12 or more hours of storage.

ORA’s extra expense argument is not valid, not only because ORA did not object to the Ivanpah PPAs back in 2008 when they were agreed to, but also because Solar Partners will pay damages for any shortfall in generation.

A payment for damages is fairly standard, according to PG&E’s filing, when any plants have a shortfall in generation:

“However, the consideration provided for in the Forbearance Agreements is consistent with other Guaranteed Energy Production (‘GEP’) damage provisions approved by the Commission, which generally base GEP payments on a rate substantially lower than the contract price. For example, in PG&E’s most recent Renewable Portfolio Standard (‘RPS’) Request for Offers (‘RFO’) form PPA, approved by the Commission in 2014, GEP damages are calculated as the difference between the current market price and the contract price, with a minimum amount of $20/MWh for GEP damages.”

So, what’s the bottom line? As PG&E put it in response to ORA’s objection, the decision results in no harm to ratepayers:

“The Forbearance Agreements also provide compensation for PG&E’s customers during this limited period.”

Ivanpah morning

As a result of the unanimous decision by the CPUC, a completely innovative technology that is one of the keys to slowing climate change was today allowed to continue to improve.

We were that generous when oil; that brand new energy technology in 1880, was first dug out of the ground and cost the 2010 equivalent of $500 a barrel. Potentially dispatchable solar technologies need to be given the same chance to develop.

Solar thermal technologies are important now that we know something about climate that we didn’t know in 1880, because, with storage, CSP can be entirely dispatchable and replace fossil energy at night.

The CPUC made the right decision.

Photos by BrightSource

Tags: , , ,

About the Author

writes at CleanTechnica, CSP-Today and Renewable Energy World.  She has also been published at Wind Energy Update, Solar Plaza, Earthtechling PV-Insider , and GreenProphet, Ecoseed, NRDC OnEarth, MatterNetwork, Celsius, EnergyNow, and Scientific American. As a former serial entrepreneur in product design, Susan brings an innovator's perspective on inventing a carbon-constrained civilization: If necessity is the mother of invention, solving climate change is the mother of all necessities! As a lover of history and sci-fi, she enjoys chronicling the strange future we are creating in these interesting times.    Follow Susan on Twitter @dotcommodity.

  • DallasTexas

    Wow talk about taxpayers and rate payers getting screwed by an industry… This and Abengoa and Solyndra and all the other smaller companies that went under. Even Solar City with its business model of collecting our income tax money.

    • Bob_Wallace

      Careful with the FUD, bub.

  • harkin

    Rate payers not harmed……but what will be the hit (additional hit that is) to taxpayers?

    • Bob_Wallace

      The output from Ivanpah will be only a tiny portion of the electricity used in a year. It will also be produced when demand is high (and the cost of electricity is higher).

      I doubt that you’d be able to see any impact on rate unless you looked out four or five decimal places. It could even be positive.

      As far as taxpayers, spending on what are largely research projects is basically money invested in a cleaner future in which energy costs less. As with any research project it might not pan out. But even failed research often points the way to solutions that do work.

      US taxpayers spend $140 billion to $242 billion a year treating health problems caused by burning coal. That’s a year after year after year expense. If we can spend a few billion dollars, even several billion dollars, and cut the money bleed caused by coal then we will have done a wise thing.

    • Taxpayers had earned about $5 billion in interest on the Loan Guarantee Program that helped investors feel comfortable lending to create new energy technologies like Ivanpah (these guarantees back up loans from private lenders) by the end of 2014 and the earnings by taxpayers are higher now.


  • eddie willers

    So….good money after bad.

  • Nikhil

    Thanks Susan for a unbiased and to the facts coverage of the issue unlike misleading articles of NYT and WSJ

    • Thank you.

    • DallasTexas

      Lol. The nyt is so conservative and anti- renewable energy. You should read around a good bit more and know your sources. I think what you meant to say is thanks for spinning an article I to the way you wanted to read it. Confirmation bias in action, bolstered by echo chamber effect of a site such as this advocacy forum.

      • Bob_Wallace

        The facts got your bloomers all bunched up, bippy?

      • Are any of my facts wrong? No, I show my sources.

        I also know for a fact that the WSJ’s reporter was given the facts and chose to write an incorrect article painting the shortfall in generation as missing the target by pretending that she was not told the target is not 100% generation for this lookback period for PG&E. Unit 1 and 3 had to be 70%, not 100% in 2014 and 2015.

        So in fact it was a 20% miss followed by a 3% miss in year two.

        Then all the other sites like the NYT, just parrot the distortions of the WSJ, due to lack of time and lack of specialized knowledge from a reporter who covers only renewable energy like I do, at various sites.

        If you think the Wall St journal isn’t an advocacy site for fossil fuels, you are an idiot. But I don’t lie about fossil fuels. The WSJ deliberately lied about this renewable project.

  • nakedChimp

    Well, for what it’s worth – I’m in favor of engineers & scientists playing with hot fluids instead of radiating materials any time.

    • Strange isn’t it, how trying something new has now become the object of such derision. But we naked chimps only got where we are today by trying something new.

  • Sorry, my mistake. PG&E’s pumped storage facility is Helms, not Hetch Hetchy.

    • Bob_Wallace

      If you aren’t familiar with it here’s the DOE Global Storage Database.

      Lots of good info.

    • Shane 2

      PV is cheaper than concentrated solar thermal generated electricity. Better to store PV electricity using pumped hydro than to store more expensive concentrated solar thermal electricity using pumped hydro. Ivanpah has no thermal storage. I smell a lemon.

      • Lou Gage

        Shane 2, How many locations in just the US have good locations for pumped hydro? CSP with storage looks like a more widely useable partial solution. Lou Gage

        • Bob_Wallace

          Lots and lots and lots of places. We have thousands of existing dams that could be used for PuHS.

          We’ve got a few thousand abandoned rock quarries and open pit mines.

          And we’ve got lots of places where the elevation changes rapidly over a short distance which creates the possibility for creating one reservoir several hundred feet over another. Just bore and sleeve a tunnel between the two.

          • Yes, there’s an interesting example of a PV plant in South America sited like that on a coastal plateau to include the onsite pumped hydro potential.

        • MattyBumpo

          Contrary to a popular notion, there are quite a few cost-effective and environmentally acceptable pumped storage sites throughout the west. The east has fewer, but still some. I disagree with Bob that there are “thousands” of viable locations, from an economic standpoint. However, there are a sufficient handful that would probably cost less than $2,000/kW, and less than $300/kWh of storage capacity (perhaps even $200/kWh of storage capacity) thus beating even the optimistic battery projections for 2020. Plus they last 75 years, maybe 100 years – many times longer than battery cells, and with far less embodied energy to manufacture. Thus, they are better for ratepayers in the long run, and better for the environment.

          • Bob_Wallace

            The US has about 80,000 existing dams and uses only 2,500 for electricity production. At least 10% of the remaining dams should have enough head and be close enough to transmission lines to be useable for PuHS.

            Abandoned rock quarries, open pit mines, and subsurface mines are other candidates. The US has around 1,000 rock quarries on federal land alone.

            Anywhere there is an abrupt change elevation over a fairly short distance is a candidate. Scoop out a reservoir high and a second one low. Drill a tunnel between the two,sleeve it, install a hybrid pump/turbine.

            Here’s a survey of potential European sites. Places where one or both reservoirs already exist. Thousands of sites.


            Cost –

            Citigroup – “We consider the PSH grid cost of $230/kWh as a future target for storage batteries to be a variable industrial infrastructure. ”

            Page 53.


            Secretary of Energy Stephen Chu told his advisory board that using pumped hydro to store electricity costs less than $100 per kilowatt-hour and is highly efficient,


            I suspect batteries will win in terms of short term storage (up to three days) but PuHS is likely to remain a viable option for the sort of deep storage we may need a few times a year.

          • Bob_Wallace

            I think we need some sophisticated modeling to tell us how much deep storage/generation we might need.

            Suppose we divided the US into a number of regions, say 20, and then looked to see how many of the 20 show a need for more than 3 days of stored power at the same time.

            Suppose we find out that at no time there are more than 2 of the 20 regions that are very low on wind, solar and other renewable inputs.

            In that case we could require all regions to install an extra 10% of generation and storage over what they need to cover 3 days of demand. And build enough transmission to move that 18 x 10% power to the two regions needing help.

            Deep storage needs disappear.

            (I have no idea what the real numbers would be. That’s why we need a Lower 48 model that uses demand and solar/wind availability data.)

          • Well, if the states like California, that want to go to 100% EV – knock out gasoline cars entirely – are successful, we will need a massive amount of new electricity generation. To avoid defaulting to extra natural gas, I think we need every renewable and every storage tech to get there.

          • Bob_Wallace

            It needs to be balanced with cost. I’m mildly skeptical in terms of thermal solar with storage being able to compete with wind/solar plus battery storage but I’m very open to data that proves otherwise.

            Thanks for doing the deep reporting on Ivanpah.

          • Well, CSP can’t be built almost everywhere like PV. It’s is so finicky about DNI, which is quite a bit rarer than basic solar insolation. So you are never going to see CSP in Massachusetts or Germany.

            It is also only feasible at utility-scale, so you’re never going to put it on a rooftop. It’s difficult to economically separate the storage aspect, so you’re never going to run a car on CSP storage the way you can run an EV on a battery storing PV or wind.

          • MattyBumpo

            Bob, your site analysis is highly simplistic. If all one needed was a modest elevation difference in order to make a cost-effective pumped storage project, there would be many, many more. The fact is that there is a huge difference in economy between a 300′ elevation site, a 700′ elevation site, and a 1,200-1,500′ elevation site. Not to mention the environmental considerations that come with adding massive pump capacity on dammed rivers that are already sensitive. Yes, new reservoirs can be “scooped out,” but that can also be expensive, so you still want the ideal topography to minimize “scooping.” That said, there are quite a few excellent sites that will beat the quoted Citigroup battery target of $230/kWh. The primary advantage that batteries at their anticipated price will be modularity and short time frame to implementation. But as noted, that advantage starts to evaporate when looking at 6-8 hours of storage, which is likely needed if we really want to leverage renewables and displace fossil for firm capacity.

  • eveee

    Nice article, Susan. Well written and thoroughly researched. Ivanpah will be a long term success and great asset.

    • JamesWimberley

      Yes. But it was still a bridge too far. Anengia has been making steady progress in CSP on a more conservative route of incremental improvements. Its financial problems come from the cost disadvantage compared to PV, not SFIK technical issues.

      • We don’t need another PV. PV is going to be cheaper than gas – for daytime. CSP is not competing with PV.

        To get to 100% renewable, we need also need another renewable tech that is dispatchable on demand. Every CSP plant troubleshot gets the engineering of being able to do that economically closer to realization.

        • Bob_Wallace

          My question is what is the likely cost of thermal solar with storage.

          PV solar is on its way to 3c/kWh. Same for wind. Storage (same day cycle) may move to 3c/kWh or less. Can thermal solar deliver for about 6c/kWh?
          What do the people developing thermal solar with storage view as the likely price once things are developed?

          • The SunShot goal is 6 cents per kWh by 2020. There is a lot of R&D – mostly into raising thermal storage temperatures to get there. Both researchers and developers have told me they do see getting to 6 cents.

            CSP has already come from 25 cents to 10 cents in a relatively few projects: just the first handful.

          • Bob_Wallace

            But that’s just for electricity delivered as generated, is it not? It doesn’t include the cost of storage.

            At 6c/kWh for directly delivered thermal solar is almost certainly dead. PV solar is already there and heading lower.

          • No, that is the total CSP+storage PPA goal – and really, only CSP that has storage is being bid these days. Most countries now either require storage with CSP bids, or pay more for dispatchability or after dark peak hours (which is only possible if storage is included)

          • Bob_Wallace

            Thanks. If they can hit 6c/kWh for stored electricity then it is probably worth continuing the work.

          • neroden

            Battery storage for daily cycling is still looking more expensive than that. 12.8 cents if you order retail now. I have no doubt that can be brought down, but I’m not sure it can reach 3 cents any time soon. I see it heading for 6 cents at utility scale, which leaves solar + batteries at 9 cents; this still leaves an opening for solar thermal with storage.

        • Shane 2

          Susan Kraemer “CSP is not competing with PV”
          This plant is competing with PV because it does not have thermal storage. I smell a lemon.

          • In general, I mean. CSP now is competing with gas – because gas delivers peaker hours after dark. These can cost 20 cents a kWh, even in the US where gas is cheap.

            The earliest CSP bids did not include thermal energy storage. 24-hour solar only kind of started with Gemasolar in Spain and the first ones in South Africa, and even the earliest CSP with storage started with only two hours, but now store up to 17 hours.

      • eveee

        Yes. Its also had a lot of flak from NIMBYs and radical fringe. And yes, IMO, there should be more thermal storage development. But heck, this is really a big breakthrough in size and performance, but still really an evolution that CSP is doing. I expect even better ones in the future. If we are smart, we will keep advancing CSP. It would myopic to let it decay until we really need the storage role in only about a decade as PV saturates.

Back to Top ↑