Ivanpah tower plant in California. Picture credits: BrightSource.

The story has what Stephen Colbert terms “truthiness” in that it reinforces the prejudice of those who think clean energy “doesn’t work.”

But is it true?

By Susan Kraemer

Ivanpah’s distinctive appearance seems to have made it an especially prominent target in the polarised political climate in the US. The innovative project was developed and funded when a Democratic majority was able to invest in new forms of clean energy with loan guarantees through the US Department of Energy.

It was designed by BrightSource Energy to deliver 1,079,232 MWh according to the National Renewable Energy Laboratory (NREL) - after a 4-year ramp-up.

Noted clean energy critic Chris Clarke, the source for a series of negative stories on CSP projects, now claims that because it didn’t generate this 12 month number in its first 8 months, it failed.

Despite some technological and weather-related challenges in its first months, Ivanpah in September was above the projected performance level, according to Jeff Holland of investor NRG Energy, which owns half of Ivanpah and is the operating partner. Developer BrightSource and google split ownership of the other half.

“The projected generation that some in the media continue to report is in fact the target goal for mature year performance, scheduled by 2018 - the four-year mark of plant operations,” he says.

So why does it take four years to ramp up?

Innovation at full scale takes time

A CSP plant essentially is both mine and power plant. It mines sunlight from a large array of mirrors that drives steam turbines. Ivanpah focuses the mirrors on a receiver atop a tower, to heat water for direct steam power production in a traditional power block below. Crescent Dunes, due online next Spring, heats molten salt.

With molten salt thermal energy storage, tower competes with coal or gas to supply evening loads. Though the first CSP used trough mirrors, tower with storage is seen as the future of the technology.

Ivanpah is the largest tower CSP in the world to date at 377 MW net - and although now a slew of 100 MW-plus tower projects are in development globally - it’s largest predecessor was Spain’s Gemasolar at just 20 MW.

“Ivanpah utilises an innovative new technology that had yet to be attempted on this size and scale,” explains Holland.

“The plant employs an incredible amount of parts and pieces and hundreds of thousands of heliostats that all have to be calibrated for maximum efficiency. Those are over and above the typical complexities associated with opening a major utility-scale power plant of any kind and those added factors were all considered when making the initial projections for the initial years of operation.”

Once the project began operations, the developers also had to make additional adjustments to reduce solar flux during standby, as part of reducing glare and avian mortality. In the first six months, 133 birds had been found with evidence of singed feathers - far fewer than the 28,000 reported - and Ivanpah has implemented ways to reduce that mortality with bird deterrents.

Truth or truthiness?

Ivanpah’s six month total generation was 173,966 MWh. But its first quarter comprised only one full month, as Chris Clarke pointed out at the time.

Ivanpah really only began full operation after the February opening, so its first quarter contained only one full month.

Ivanpah’s failure to achieve in 8 months the 12 month figure projected for 2018 may “concern” those obsessed with its failure, but the project owners are operating with long-term goals in mind.

Holland earlier commented to CSP Today that lower than average DNI this year impacted generation, but was not concerned as they had accounted for varying weather over time as well as a ramp-up involving ongoing R&D.

NRG Energy owns more renewable generation than all but two US companies, and Holland was a bit taken aback by the level of negative attention Ivanpah had attracted by August with exaggerated avian mortality stories.

Ivanpah is actually more environmentally sensitive than most other energy-producing sites because ground is not disturbed, so wildlife like kit foxes and snakes have the run of the land, and a nursery run by 25 biologists protects baby tortoises found on site.

“It's remarkable what that project has had to endure and continues to, when you understand the incredible efforts and dedication the people on-site have undertaken to preserve the flora and fauna,” Holland said. “It's become a lightning rod political issue for news, which is unfortunate because of the enormous positves that Ivanpah brings from a clean energy standpoint. NRG believes in a clean and sustainable energy future and the renewable technology at Ivanpah supports this goal.”

Ramp-up period not unlike other new CSP

How does Ivanpah compare with another CSP project that has escaped this sort of obsessive scrutiny?

The 250 MW Genesis project, using trough CSP technology, with planned generation of 580,000 MWh annually, came quietly online in the third quarter of 2013. Trough CSP has historical precedents at this size.

Notwithstanding Ivanpah’s unique tower technology, its gradual increase in production is not unlike a similar ramp-up at Genesis.

According to EIA figures, Genesis generated 7,029 MWh in its first quarter, then 66,427 MWh in its second, and 141,939 MWh in its third quarter. Ivanpah generated 40,074 MWh in its first quarter and 133,892 MWh in its second quarter.

Genesis began in 2013, and Ivanpah began in 2014. (Genesis operated for two quarters before announcing its opening in April of 2014.)

DNI varies as much as 20% annually

One factor that the CSP industry needs to pay heed to is variability in Direct Normal Irradiance (DNI). Good DNI is essential for CSP and it’s highest in clear high desert air.

But it is not a static metric that is always the same in a particular region. It can actually be 20% higher, or 20% lower than average in any given year, depending on weather, pollution and dust.
“BrightSource really was ahead of its time,” says Gwendalyn Bender, Product Manager for Assessment Services at Vaisala, which recently acquired 3TIER, the renewable energy consulting firm that provided the solar resource assessment for Ivanpah.

“Back in 2009 they reached out to us to do a long-term study with on-site ground measurements to complement a 10 year-plus record of variability based on satellite records. So they really did do their due diligence at the time, to look at the variability that is possible, and integrate that into their performance plan.”

Wildfires affect CSP much more than PV

The key variable for PV is Global Horizontal Irradiance (GHI). But dust particles don’t affect solar PV production as much due to the diffusion effect.

“With all the smoke and everything in the atmosphere, fire definitely affects DNI,” says Bender. “DNI varies more than GHI from year-to-year, and if there is dust or pollution in the atmosphere you will see the effect sometimes even 10 times more than in GHI.”

Vaisala’s maps indicate above and below normal DNI with reds or blues. In 2013, California was having a better-than-average year for DNI, but then had the Yosemite Rim Fire, which shows up by contrast.

“Last year you saw a blue streak on the map where the wildfire and the smoke from the wildfire was occurring,” Bender explains. “What you see from something like a wildfire is a reduction in DNI for a particular area. You don't see that as clearly this year because it was already all blue.”

She says DNI levels in the summer of 2014 were 10% below average, from both weather and wildfire debris. CSP generation took a corresponding dip in July that continued into August.

But renewable generators understand variability, and plan generation estimates over project lifetime accordingly.

Despite the calibration of groundbreaking technology at this scale, additional testing to reduce solar flux, and a summer of DNI that has been 10% below normal in the region, Ivanpah’s output doesn’t indicate that tower CSP has a technical problem.