Here’s why residential solar is down more than expected in Q3 2017

residential solar installation

Despite more than half of U.S states now being at grid parity — meaning the levelized cost of energy is below electricity bill savings in year 1 of system life — the U.S. residential segment posted its lowest solar installation total since the first quarter of 2015, according to the latest U.S. Solar Market Insight report, GTM Research and the Solar Energy Industries Association (SEIA).

The residential PV sector fell 10 percent quarter-over-quarter. Declining growth is driven by weakness in California and major Northeast markets, which continue to feel the impact of pull-back from national providers.

The report attributes the slowdown to two key factors: persistent nationwide customer acquisition challenges and a pivot by major solar installers that are pursuing profitable sales channels over growth. This has been particularly acute in mature markets that account for the majority of installation volumes.

Several markets, however, experienced record quarters for the residential solar segment. These include New Mexico, Washington D.C., Virginia and Idaho. Meanwhile, emerging markets, such as Florida and Pennsylvania, are expected to surpass 50 MW of residential capacity for the first time ever this year.

“The year 2017 has been unconventional for solar in the sense that utility and residential PV, which have historically been the market’s major growth segments, are actually expected to decline in 2017,” said GTM Research Solar Analyst Austin Perea. “For utility PV this is largely a function of comparing the record-breaking ITC demand-pull in effect of 2016 to more modest build-out in 2017, while significant customer acquisition issues remain a challenge for residential solar. Conversely, non-residential solar, the smallest and most historically beleaguered sector, is expected to grow in 2017 in large part due to robust community solar build-out and regulatory demand pull-in across major state markets.”

— Solar Builder magazine

7 solar install stats to know from third quarter 2017 report from GTM Research, SEIA

PV installations 2017

In their latest U.S. Solar Market Insight report, GTM Research and the Solar Energy Industries Association (SEIA) said prices rose due to a tight global supply of modules and uncertainty around the Section 201 trade case now being weighed at the White House. What effect did this have on installations across solar segments? These numbers jumped out at us.

2,031 MW

That’s the number of megawatts of PV installed in the U.S. in the third quarter of the year — the eighth consecutive quarter that the solar industry added more than 2 GW. The capacity additions represent a 51 percent drop year over year.

10 percent.

That’s how much the residential PV sector fell quarter-over-quarter. Declining growth is driven by weakness in California and major Northeast markets, which continue to feel the impact of pull-back from national providers. More on that here.

22 percent.

That’s how much the industry is down compared to this point last year, which is in line with the expected 21% decline for all of 2016 vs all of 2017.

11.8 GWdc.

That’s how much new PV GTM Research estimates will come online in 2017, down 22% from a record-breaking 2016. The forecast has been adjusted downward from 12.4 GW last quarter to reflect continued challenges in the residential market and a push back in utility-scale completion timelines due to uncertainties surrounding the trade case.

22 percent.

Get this: That’s how much non-residential grew year-over-year, installing 481 MW in Q3. Non-residential consists of commercial and industrial businesses that install solar, nonprofits, and community solar programs. California, Massachusetts and New York all posted strong quarters while Minnesota had its largest quarter ever due to its robust community solar program. Nationwide, community solar capacity is on track to grow by more than 50 percent year-over-year.

utility scale solar installation

51 percent.

That was the utility-scale segment’s piece of of the quarter’s installed capacity pie. The utility-scale segment was led by Nevada, North Carolina and Texas. In fact, Texas installed more solar in the third quarter of this year than the state installed in the entirety of 2015. Meanwhile, emerging markets in the Southeast, including Florida, Mississippi, and South Carolina all had strong quarters and are forecast to install more solar in 2017 than any year previously.

4 GW.

Encouragingly, that’s how much utility-scale PV is currently under construction across the nation, and GTM Research forecasts an additional 3.9 GW will come on-line by the end of the year. This would make 2017 the second largest year ever for solar installations behind only the record-shattering 2016.

“Looking forward to 2018 and beyond, both Section 201 remedies and corporate tax reform present considerable downside risk to the industry’s base-case forecasts. However, at present, neither issue will be incorporated into GTM Research’s existing outlook until President Trump issues a formal decision on Section 201 trade remedies and the U.S. Congress votes on corporate tax reform legislation.”

— Solar Builder magazine

GTM: Sunrun tops SolarCity as lease sales leader, and more residential solar sales stats

Sunrun logo

2017 isn’t quite the banner year for solar that 2016 was. Aside from the trade case drama, it has also been a tumultuous year for the big names in the residential solar space. Three of the largest installers, including NRG Home Solar, Sungevity and Direct Energy Solar, have gone bankrupt or exited the residential sector. Since struggling SolarCity was acquired by Tesla, its residential business has dwindled. But Sunrun, who has seen moderate and consistent deployment growth over the last few years, has worked to fill the gap and serves as a prime example that struggles in the residential solar industry may stem from company-specific failings rather than industry-wide trends. GTM Research saw this coming, and relayed the following stats on the residential solar sector.

sunrun solar city leases

As a residential lease and PPA provider, Q3 earnings presentations indicate that Sunrun has already surpassed SolarCity based on capacity financed so far in 2017. Through the first half of the year, Sunrun narrowly missed the top spot in the TPO market with 27% market share, just behind SolarCity’s 31% share and up considerably from its 18% share in 2016. That difference of 4% market share between SolarCity and Sunrun equated to just 19 MW over two quarters. And in Q3, Sunrun financed 80 MW of systems, while SolarCity financed no more than 59 MW (a ceiling, as some of SolarCity’s systems were from its commercial business), a difference in Sunrun’s favor of more than 20 MW.

Of course, there are other ways to look at the market outside of who is financing systems. Much of SolarCity’s fall as a top residential financier has been due to its deliberate pivot away from TPO financing in order to increase its cash position. Today, nearly half of SolarCity’s systems are sold for cash or loans, and this pivot is inextricably linked to loan provider Mosaic’s prominent position in the loan market.

But looking at the residential market by total deployments (including leases, PPAs, loans and cash sales), GTM says Sunrun likely surpassed SolarCity as the leader in the space for the third quarter of 2017. According to the U.S. Residential Solar Finance Update, SolarCity deployed 233 MW of residential solar in H1 2017, Sunrun deployed 148 MW, and Vivint Solar deployed 93 MW. Yet in Q3 2017, these companies deployed 109 MW, 90 MW, and 47 MW, respectively (SolarCity’s 109 MW includes its commercial business).

So, if 18% or more of SolarCity’s Q3 installations were in its commercial business (which is reasonable given SolarCity’s historic channel mix), then Sunrun would have narrowly out-installed SolarCity in the quarter.

New sales channels and finance offerings

sunrun solar city

Both SolarCity and Vivint have endured high customer acquisition costs as mature markets have become oversaturated, and the companies have been forced to scale back operations in unprofitable markets. Specifically, SolarCity has dropped its door to door sales channel and instead is focusing on acquiring customers through digital leads. Vivint Solar, which has traditionally relied primarily on door to door sales, has added retail sales to its mix. And while these customer acquisition strategy changes are aimed to bring costs down in the long term, the slow-moving transition to these new strategies has had a short term effect of increasing costs and decreasing sales.

Equally important, both SolarCity and Vivint Solar have made concerted efforts to increase cash and loans sales as a portion of their product mixes.

While the companies make better margins off their TPO products, years of selling leases and PPAs (where the companies receive payment from the customer over a 20 year term) have left both companies in dire need of cash in the near term. Cash and loan sales allow the installers to realize immediate payment for systems they install. But even this change comes at a cost. SolarCity and Vivint Solar employ salespeople who have been selling leases and PPAs for more than 10 and 5 years, respectively. The transition to selling loans has been difficult on sales teams that are forced to change their long-honed pitches, contributing to the sales declines by these companies.

But as nearly every other large national installation company has struggled to grow this year, Sunrun is a standout as its growth has outpaced the market. Unlike its largest competitors, Sunrun has seen customer acquisition costs come down in recent quarters. And unlike SolarCity and Vivint Solar, Sunrun services the market both through its direct installation business as well as with Sunrun leases and PPAs delivered through its dealer network. By utilizing a dealer network to deploy systems, Sunrun is able to grow as the long tail of installers in its network grow as well. And while not all long tail installers are growing, Sunrun’s stringent vetting of installer partners weeds out the weaker installation companies who are more likely to go in and out of business with market boom and bust cycles.

Fact: People want solar energy. Here are some stats on why

Is the residential solar financier shakeup here to stay?

As SolarCity and Vivint Solar have deliberately scaled back operations and moved away from employing a strictly vertically-integrated installer and financier model, Sunrun has jumped on the opportunity. Unlike its competitors, Sunrun continues to primarily sell TPO systems through its direct and installer network businesses.

But recent success for Sunrun does not guarantee continued success. While Sunrun is now the leading TPO financier in the residential solar market, questions remain as to the size of that addressable market. As the residential market grows into the future, GTM Research expects the TPO market to stay relatively flat through 2022, putting a ceiling on the market that Sunrun can address. The current transition of the market away from TPO, which, according to GTM Research, will make up just 37% of the residential market in 2017 as compared to 53% in 2016, is primarily due to what leading installers are choosing to sell.

But there is downside risk to the size of the addressable TPO market. As residential system costs continue to decline, consumer-driven demand for TPO financing could become a prevailing force squeezing that market, leaving Sunrun behind the curve. There is certainly ample opportunity for Sunrun to increase its market share with its leases and PPAs, though the company has little room for error in a market with a low ceiling.

— Solar Builder magazine

Tariff impact on the U.S. solar market in 7 charts from GTM Research

GTM research solar tariff prediction

The imposition of tariffs on imported solar cells and modules is more a matter of when than if at this point considering the Administration in charge, so GTM Research has been furiously crunching the numbers to understand the fallout across the U.S. solar market based on different tariff scenarios. Here’s a glimpse into that crystalline ball.

Reminder: The ITC will vote on Oct. 31 on a remedy recommendation, submit that recommendation to the White House on Nov. 13, and then Trump will have until January to make a decision. That decision will be imposed 14 days later.

A note from the GTM team: “Our team has been working on this analysis for over a month, so our scenarios (which range from $0.10/W-$0.40/W cell tariffs, in increments of $0.10/W) don’t perfectly match with either Suniva or SolarWorld’s proposed remedies. But SolarWorld’s request aligns closely with our $0.30/W cell tariff, and Suniva’s request is close to our $0.40/W cell tariff scenario.”

Current situation

GTM solar tariff chart_global selling average

Tariff risk has caused module prices to increase, a phenomenon unique to the U.S. While cost reductions in other parts of the system make up some of the difference, the cost to install solar has increased for the first time in ages.

GTM solar tariff chart 2_EPC turnkey pricing

So what comes next?

GTM solar tariff chart 3_tariff free capacity

If tariffs are imposed, GTM estimates that there will be nearly five GW of solar capacity that is not subject to tariffs, either because it is not subject to the scope of the petition (i.e. thin film) or because both the cells and modules are manufactured in the U.S., Korea, Singapore, Canada or Australia, all of which may be exempt. In addition, over 2 GW of modules have already been procured for 2018 projects, which will temporarily dampen the tariffs’ impact on demand.

Those five gigawatts won’t be nearly enough to sustain the market, which is otherwise expected to reach nearly 11 GW in 2017, rising to over 16 GW by 2022.

Tariff impact demand

GTM solar tariff chart 4_PV capacity

GTM estimates that the net impact to its base forecast could range from just 9 percent under a 10₵/W tariff to 48 percent under a 40₵/W tariff. The biggest impacts would be in the utility-scale solar sector, which is most sensitive to price in-creases, while the residential sector would be the most resilient.

GTM solar tariff chart 5_residential solar impact

Every segment, in every state, will be unique. In the residential sector, the biggest volume impacts would be felt in the largest state markets, but nascent states that have just begun to develop vibrant residential solar sectors could disappear almost entirely.

The utility-scale market would be most sensitive because two-thirds of the project pipeline is driven by solar’s razor-thin economic competitiveness with other generation sources.

GTM solar tariff chart 6_utility pipeline

But even the utility solar market could weather a 10₵/W cell tariff with relatively minimal disruption – just over 10% by our estimate.

GTM solar tariff chart 7_utility installs under tariff scenarios

— Solar Builder magazine

Module Evolution: What big-time PV improvements will boost panel efficiency?

Light Bulb Illustration

The solar industry is forever in need of constant innovation and scientific breakthrough. Even now, with PV capacity reaching record highs, prices continuing to fall and efficiencies inching up, more innovation is needed. A lot more. (Not to mention whatever fallout is felt in the module market following the Section 201 trade remedy.)

At Intersolar North America in July, Martin Keller, director of the National Renewable Energy Laboratory (NREL), told attendees that if new materials and new production methods don’t hit the market, solar will never make the impact we all think possible as a distributed energy source.

“What are some new materials we can combine with new manufacturing technologies?” he asks. “If you are really serious about manufacturing on a global scale, we need new methods and new materials.”

If you are reading this, we assume you meet Keller’s “really serious” criteria and thus would like an update on some of those new methods and materials on the horizon. As far off as Keller made 2017 technology seem from where it needs to be, we think there are enough smart people working on stuff right now that this challenge will be met.

Silicon Successors

NREL is at the forefront of renewable research and pushing innovation, and scientists there have developed a new perovskite ink with a long processing window that allows the scalable production of perovskite thin films for high-efficiency solar cells. Keller was excited about it and pointed to a chart that showed a severe uptick in efficiency compared to today’s cells.

The catch is perovskite solar cells have yet to move beyond the laboratory. The crystalline structure of perovskites must be carefully grown upon a substrate, which is normally done by laboratory-scale spin coating — a technology that can’t be scaled to large-scale manufacturing at this time.

“It’s years out from production, but you can see the increase in efficiency is a very steep slope, and then combine that with new technologies like inks and spray ons,” Keller says.

Over at Penn State, researchers are testing a prototype of a new concentrating photovoltaic (CPV) system with embedded microtracking that can produce over 50 percent more energy per day than standard silicon solar cells. CPV focuses sunlight onto smaller but much more efficient solar cells, like those used on satellites, to enable overall efficiencies of 35 to 40 percent. Current CPV systems are large — the size of billboards — and have to rotate to track the sun during the day. These systems work well in open fields with abundant space and lots of direct sun.

“What we’re trying to do is create a high-efficiency CPV system in the form factor of a traditional silicon solar panel,” says Chris Giebink, a Charles K. Etner assistant professor of Electrical Engineering at Penn State.

To do this, the researchers embed tiny multi-junction solar cells, roughly half a millimeter square, into a sheet of glass that slides between a pair of plastic lenslet arrays. The whole arrangement is about 2 centimeters thick and tracking is done by sliding the sheet of solar cells laterally between the lenslet array while the panel remains fixed on the roof. An entire day’s worth of tracking requires about one centimeter of movement.

“Our goal in these recent experiments was to demonstrate the technical feasibility of such a system,” says Giebink. “We put together a prototype with a single microcell and a pair of lenses that concentrated sunlight more than 600 times, took it outdoors and had it automatically track the sun over the course of an entire day.”

The researchers report that the CPV system reached 30 percent efficiency, in contrast to the 17 percent efficiency of the silicon cell. All together over the entire day, the CPV system produced 54 percent more energy than the silicon and could have reached 73 percent if microcell heating from the intense sunlight were avoided.

But (there is always a but) Giebink noted that major challenges still lie ahead in scaling the system to larger areas and proving that it can operate reliably over the long term. Insert sad emoji here.

While we wait for those new markets to scale and develop, there are a bunch of intriguing options that could boost efficiencies and bridge the gap.

Production Disruption

Rayton Solar

Rayton Solar wants to supplant the very way we cut silicon in the first place — a technique that hasn’t changed much since its inception in the 1950s. Cutting silicon with a diamond saw leads to a significant amount of sawdust because the process wasn’t originally concerned with reducing waste for large-scale production.

“We developed a process using ion implantation to cut our very thin pieces of silicon, and there is zero sawdust in the process, so it allows us to increase the yield of the raw silicon and get a 60 percent reduction in the cost to make a solar panel,” says Rayton Solar CEO Andrew Yakub.

Phoenix Nuclear Labs (PNL) has signed a long-term agreement to be the exclusive supplier of high-current proton accelerators to Rayton Solar to produce low cost, high efficiency solar panels. Under the terms of the agreement, PNL will deliver the first system to Rayton at the end of 2017, followed by several additional units in 2018 and 2019.

The Rayton process utilizes high current ion beams produced by the PNL technology to cleave thin layers of silicon with zero waste. The process uses 50-100 times less silicon than the traditional method. Because of this, Rayton can also use a higher quality silicon that is about 10 times as expensive.

“We are capable of making up to 100 times as many solar panels with the same amount of silicon that our competitors use to make just one panel,” Yakub says.

In a less radical direction, mono passivated emitter rear cells (PERC) have efficiency seekers excited, and advancements keep happening every day. Silicor Materials says that, in its first ever attempt, it has produced p-type mono PERC cells at approximately 20 percent efficiency, using 100 percent of its standard silicon feedstock. Silicor hopes its technology for manufacturing solar grade silicon provides the solar market with a simple solution to manufacturing the highest quality, highest efficiency solar cells of the future at a substantially lower cost than all other solar grade silicon manufacturing technologies on the market.

Sol Voltaics has taken a big step toward commercializing a new efficiency-boosting solar technology. Using its proprietary Aerotaxy process to manufacture PV nanowires, its SolFilm solution could boost solar module performance up to 50 percent at a low cost.

SolFilm consists of billions of gallium arsenide (GaAs) nanowires oriented facing the sun. The nanowires, each of which is a complete solar cell, convert high-energy sunlight directly into power. Gallium arsenide, previously seen in space and concentrated solar projects, has long held great potential for the mainstream solar industry, but its high fabrication costs have prevented economical fabrication of large solar panels.

Manufacturing nanowires with Aerotaxy dramatically reduces the required amount of GaAs and removes the need for a crystalline support wafer, significantly lowering material costs.

“The nanowires are grown such that the top and bottom of the wire have opposite doping profiles. This makes each nanowire a fully functional solar cell, with a pn junction along the length of the wire,” states Erik Smith, CEO of Sol Voltaics. “Whether used by module manufacturers as a single-junction, high-efficiency, low-cost solution or as a boosting technology, we believe SolFilm will usher in a new age of solar power efficiencies.”

Sol Voltaics just closed a record funding round of $21.3 million (following a $17 million investment last year). The new funding will be used to accelerate commercialization of the technology.

You Down with BIPV?

Forward Labs Solar Roof

This is the part of the modules section when we throw an obligatory mention to Tesla and its new, mysterious Solar Roof. Building-integrated PV tiles are not new, but like the cool kid who started wearing bell bottoms to school, Elon Musk has made them trendy again. Any casual conversation I have about the solar industry outside the office always leads to the layperson asking about Tesla’s Solar Roof. So, word is out.

The buzz for the industry is certainly a good thing, but those everyday homeowners might not get the best bang for their buck going with the Tesla Solar Roof. Online solar marketplace EnergySage ran numbers on comparative systems for a 3,000-sq-ft home in Southern California with a $200 monthly electric bill, as an example, and the results speak for themselves:

  • Standard PV system: $26,030; 13,000 kWh annual production
  • Tesla Solar Roof: $50,900; 10,000 kWh annual production

The real hook of the solar roof is how it replaces the roof itself. But if you add in a $20,000 cost for a roof replacement as EnergySage did (based on a Consumer Reports estimate of such a job for that house size), the non-solar roof is still a better value.

Put more simply, GTM Research determined that Tesla Solar Roofs produce about 6 watts per square foot, whereas a high-efficiency module would produce 19 watts per square foot. There is also a potential hang up with applying for ITC credits because not all of the shingles being installed will be solar shingles.

Anyway, we wouldn’t bet against Tesla making this concept happen as the costs become more competitive over time, but a bit more quietly, Palo Alto-based startup Forward Labs entered this space at the same time as Tesla, claiming to be 33 percent cheaper, more efficient and easier to install — 19 watts per square foot of energy density at about $3.25 per watt, installed in two to three days.

“The way we achieved such fantastic cost savings was fairly simple,” Zach Taylor, CEO and product architect of Forward Labs. “We use more affordable materials than our competitors and employ standard manufacturing processes. The roof’s installation process is simple and quick — we can install our system in half the time that other companies can. The benefit to homeowners is a return on their investment that cuts the usual solar payback time in half.”

Forward Labs uses a proprietary five-layer construction. A robust glass panel sits atop an optical layer, which cloaks the underlying black monocrystalline solar cells and enables eight possible color choices. These top layers are embedded over a galvanized metal form-factor that appears nearly identical to the non-solar portions of the roof.

“The colors of solar roofing products have always been muted or limited in choice for the sake of energy production,” says Reid Anthony, former CEO and president of precision optics company Kowa American. “Forward’s embedded optics have overcome these challenges, giving homeowners the freedom to have a solar roof in some of the most desired colors, without increasing the cost or sacrificing energy production. It’s a game-changer for both consumers and the solar industry.”

The solar roof not only weighs the same as a composite shingle roof, its sleek design also vents cool air under the solar cell layer, keeping operating temperatures down while maximizing cell efficiency.

“Although most of the technology has been developed in-house, we’re proud to have developed Forward’s panels with high-quality materials from LG, Valspar and other Fortune 500 companies,” says Taylor. “This will enable Forward to quickly establish a strong network of key supply chain partners.”

Tweaks on the Traditional

panasonic

Current technology still offers a ton of potential, especially with tweaks to traditional panel architecture. Here are four recent developments.

1. Maxim Technology, which we initially reported on to start the year in our Innovations Issue, is gaining momentum with its module optimization technology — a chip that is installed directly into the PV module instead of a diode. The installer can simply wire this system with a string inverter as they normally might and achieve full optimization, MPPT and rapid shutdown compliance.

The Maxim technology, over time, could change the value of high-efficiency modules too. Certain mono PERC modules, for example, are prone to hotspots, which can counteract their added efficiency value. Incorporating cell-level optimization would remove that issue.

2. The two leading thin-film solar manufacturers, First Solar and Solar Frontier, represent a combined manufacturing capacity of 4 GW. While they do not pose a short-term challenge to crystalline silicon players’ market dominance, ongoing innovations will ensure thin-film remains a significant player, according to Lux Research.

Of the two, First Solar is far bigger, with expertise in utility-scale systems and a new large-format module design that will help maintain its GW-scale presence in utility-scale systems, as deployment grows in emerging markets. Solar Frontier has gradually diversified its business away from its home market of Japan and is making steps toward a rooftop BIPV product.

First Solar’s further growth hinges on plant-wide adoption of its Series 6 module and achieving systems costs below $1 per watt. Solar Frontier’s future rests on its ability to move its success in the lab to commercial production, and a partnership with a storage provider to integrate a lithium-ion battery option with its residential systems.

3. In the add-on category there is PLANT PV’s new Silver-on-Aluminum paste. The goal here is providing a 1 percent increase in relative power output for c-Si solar cells via easy implementation with no added investment cost for cell producers. Silver-on-Aluminum paste provides cell manufacturers with the ability to print the paste directly onto dried aluminum film, allowing them to cover the entire back of the wafer with aluminum paste and obtain the beneficial passivation of a continuous aluminum back-surface field.

“For 20 years the industry has had to accept an efficiency loss from printing silver bus bars directly onto solar cells,” stated Craig Peters, CEO of PLANT PV. “Our Silver-on-Aluminum paste has been developed to directly address this problem and enable cell producers to eliminate these unnecessary efficiency losses in all conventional solar cells today.”

4. Incremental efficiency improvements continue from the traditional sources as well. Panasonic Corp. achieved a new leading output temperature coefficient for mass-produced silicon photovoltaic modules, at -0.258 percent /°C. This improves on the previous temperature coefficient by 0.032 points at the mass production level, highlighting the positive temperature characteristics of heterojunction solar cells and further improving Panasonic’s unique heterojunction technology.

Panasonic HIT modules, which boast an improved output temperature coefficient, will nearly halve the decline in the conversion efficiency, significantly increasing performance in high temperature settings.

Chris Crowell is managing editor of Solar Builder.

 

— Solar Builder magazine