Large-Scale PV Operations and Maintenance
Inside this Article
Well-planned O&M helps mitigate 30 years of risk.
Using solar power is essentially prepaying for energy—or prepaying for fuel, in the parlance of utilities—30 years in advance. PV requires harnessing the sun’s power today and every day for the next 30 years. On a missed operational day, solar radiation is dissipated as heat. This is a lost opportunity, like pouring gasoline onto the ground. Even one day of lost energy is wasted potential. This is why the operations and maintenance (O&M) aspect of a PV system is so important. Nevertheless, it is often overlooked.
Despite the fact that PV systems are low maintenance, they are not completely maintenance free. Solar is an excellent, reliable source of power, but PV systems must be diligently observed and operated to capture as much of the convertible power as possible. In this way, owner/operators can achieve the benefit necessary to pay off their investment. Many developers try to convince investors that panels can be forgotten once they are turned on. This oversimplification underestimates O&M and allows developers to reduce cost assumptions to boost profitability.
These assumptions are hard to challenge since independent engineers typically do not have 30 years of failure history to refer to. According to Sarah Disch, director of operations for Fotowatio Renewable Ventures, “The cost of corrective maintenance is largely unknown, and it is not clear that the relatively small operating budgets established during project financing for 20-plus years will be sufficient.” Because the future costs for unscheduled maintenance due to failures are an unknown, all parties are reluctant to predict on the low end of probability. Sometimes the resulting high-end prediction creates an O&M budget that stops a potential system from being built.
O&M for PV systems is managed differently from plant construction. Optimal operations must strike a balance between the most productive system and the one with the lowest costs. Operational decisions are challenged by the financial structure, the maturity of the industry and economic returns. Many maintenance providers offer O&M as a guaranteed service to encourage the sales process. Operations are also about the organization and management of contract obligations. The engineering, procurement and construction (EPC) contractors are not necessarily equipped to manage multiple long-term contracts with layers of guarantees.
Operations are often like death by a thousand (paper) cuts. Maintaining a plant has less to do with the actual work in the field and more to do with data analysis, potential issues and contract management. Large systems are subject to layers of obligations by the project sponsor, EPC contractor, subcontracted parties, insurance provider, utility, site host and so on. These systems can have thousands of pages of project documents. Understanding each of these documents is essential to ensure the owner is not overpaying and is getting the services promised. While O&M contracts are individually negotiated, providers may not recall the specific nuances of a deal, tending instead to treat all operations similarly. It is just as important for the service provider to be well versed in these contracts. The owner may ask for what seems like prudent service, but it may not be covered in the service contract. These service contracts are subject to myriads of lawyers and red pens. They may not be nearly as boilerplate as they appear.
O&M is a mindset that requires the strategic evaluation of a system to know how to predict the economic and contractual impact against less-than-optimal performance. Understanding the expected costs means budgeting accordingly. Once the system is in operation, understanding the financing structure is crucial. The most technically prudent solution may not be the best for the project. Finally, O&M is a daily exercise. It cannot necessarily be scheduled quarter-byquarter into the future. An operations team must understand the overall system performance, observe trends and help prevent problems or solve them immediately after they occur.
The solar O&M market is mostly made up of integrators who were forced to provide operational guarantees to sell products over the past approximately 5 years. This creates some inherent challenges. For example, EPC contractors are still being obliged to provide commitments to system performance that are not part of their focus. These large organizations have small groups focused on operations attempting to fulfill commitments made by a sales team years earlier. It can be difficult for PV system owners to get the service they paid for from the large construction company.
As the scale and diversity of solar plants continues to grow, large contractors may not be able to effectively staff every site. This growing need for O&M services may represent an opportunity for regional integrators. A local provider, for example, could provide operations services, either to the EPC contractor or system owner.
ECONOMICS OF MAINTENANCE
Maintenance is divided into two parts: scheduled and unscheduled. Scheduled maintenance typically comprises a list of required services by equipment manufacturers outlined in the owner’s manual. This information can be vague, depending on the manufacturer. It can be very helpful to discuss the intention of the scheduled maintenance directly with the manufacturer. Unscheduled maintenance involves fixing problems once they occur. These are not two distinct categories, but rather they comprise a balance of needs. For instance, a lack of scheduled maintenance increases the amount and typically the cost of the unscheduled maintenance. As President Adam Burstein of Next Phase Solar says: “The single biggest contributor to poor system performance is poor preventive maintenance. Many times, costly system outages could be avoided with periodic inspection of inverters, combiner boxes, array wiring and so on.”
Operational performance is complicated by the fact that many large-scale systems use a tax equity flip structure. This financial structure gives most of the benefit to the equity investor during the investment tax credit recapture period— about 6 years—and then flips the equity to the developer for the remaining life of the system, giving the upside to the developer. Therefore, running a cost-benefit analysis is not as simple as comparing the expected production benefit and the cost of the work. An increase in the production during the first 6 years simply shortens the time period until the flip. This increase in production and revenue mostly belongs to the tax equity investor. Operational expenses give the developer access to the upside earlier and are more about the time value of money than a direct increase in revenue.
Whatever the project’s financial structure, system production and availability are crucial to the economics of maintenance.
Production. Economic returns are based on percentage points of production, the energy harvested by the system. This may sound counter to the flip tax structure. However, developers face two challenges. First, the project revenue projections need to be high enough to ensure that everyone has satisfactory returns for the life of the asset. Second, these production numbers must be conservative enough to be attainable. Developers need to hit this sweet spot to attract capital for the next deal or add value for selling the company or project pipeline.
Solar projects are obviously extremely sensitive to performance. Usually total operational expense is about 10% of revenue. If it goes up by 5%, production also needs to go up a few percentage points to get the project back on track. When determining the viability of a project, production is one of the top three drivers, along with power purchase agreement pricing and installation costs. A higher-revenue contract makes production even more sensitive. Despite the fact that production should be the number-one driver for PV site locations, projects are being driven instead by the highestrevenue contracts.
Availability. In the context of O&M, the percentage of time that a system is operating properly is referred to as its availability. System availability is critical to the overall physical and economic performance of the plant. After all, a photon makes just one trip from the sun. The most efficient system is not useful if it is not converting that photon to power. In fact, most operational challenges are about on-line availability, not percentage point discrepancies attributed to the system design.
Availability in the solar service industry has suffered from absurdly low projections. An availability of 98% for a 10 MW facility means that every inverter is off line for more than 7 days every year. This is not acceptable. Scheduled and unscheduled maintenance for monitored power electronics should be able to achieve 99.9% availability. Constant monitoring is essential to ensure that the system is operational. When the system has a failure, there must be a clear recovery mechanism to get it back on line. However, as solar projects have a widening geographic footprint, the ability to access the system becomes more challenging.
A comprehensive risk management program involves identifying what can go wrong, its probability and its consequences. Operations require the owner to understand the performance of the system and identify risks to quickly prioritize, mitigate or assume them. Maintaining optimal performance involves clearly and quickly translating the status of a system to its impact on project contracts and, ultimately, the investment. The principal modes of failure and exposure can generally be identified. However, the probability of failure is either unknown or guarded in an emerging market.
Even more challenging, planning for unscheduled maintenance requires making an assumption about the probability of failure. Good equipment manufacturers have data on direct failure modes and some idea of expected frequency. This information is generally protected, because some manufacturers do not want to show weakness in their product, especially as they compete as commodities. However, it is the indirect failure mode that is important. For example, an inverter manufacturer knows that heat can affect switches, resistors, breakers and so on—but it is not known how quickly an inverter placed in the high desert will fail, or what the impact of that failure will be. It is just accepted that it is higher. There is not a lot of failure data available for developers, because there is not a lot of project history. In addition, OEMs are reluctant to share site-specific incidents.
Modules and inverters are the biggest sources of risk and the resulting consequences. Initially, the trouble often manifests as a poor installation technique or incomplete commissioning, but long-term system degradation also plays a large role.
System failures can come in many different forms with a variety of consequences—some not always readily apparent. Failure can be somewhat counterintuitive. Integrators on large utility-scale projects often offer strong performance guarantees that cover all expected revenue. These guarantees may protect the system in the event of a failure. However, some of these guarantees require reliable data. Therefore, it may be more important to keep the communications system operational than the actual PV system. Counterintuitively, failure of an inexpensive data-monitoring component can be more important than an expensive piece of power electronics.
Inverters. Inverter failures are the most common causes of system downtime. Inverters are sophisticated power electronics that will fail; it is only a matter of when. These incidents can be as catastrophic as a failure of insulated-gate bipolar transistors or capacitors. However, inverters can trip from hundreds of internal conditions—and these trips can have consequences similar to a failure. It can take days for a technician to get to the site and reset the inverter. In the end, either a catastrophic failure or an easy-to-fix trip can have the same downtime effect.
Modules. Module failures also occur but are challenging to identify. Typically, modules fail either physically or through incremental reductions in performance. Physical failures usually appear soon after installation. A few modules, depending on the manufacturer, may break within the first months of installation. David Vincent, project development manager for Conergy’s Projects Group elaborates: “First Solar accounts for 2% breakage, though we are seeing far less than 1% on most of our projects over the first year. Its formal time period for replacing broken modules under warranty is 5 years. Since a broken module produces about the same energy as intact ones, performance is not sacrificed.”
In large fields, there can also be some failure due to delamination. However, it is difficult to determine if a single module has a modest reduction in production or is a complete failure. Even the complete loss of a single module can be difficult to identify. It is always a good idea to wear polarized sunglasses to see any discolorations that may indicate the beginning of delamination. Module failures are rare but must be treated carefully within the guidelines of the warranty.
Trackers. Trackers have two main modes of failure: alignment and the system’s programmable logic controller (PLC). Tracker alignment is essential to avoid stress on the system that can increase the amount of current draw on the tracker motor and shorten its life. This can also cause a reduction in production, depending on the tracker’s position when it fails. A tracker PLC could fail due to the algorithm or system component used to accurately track the sun. Both of these failures reduce production. The benefit of a simple design is that most problems can be easily fixed. However, performance guarantees or warranty obligations may be voided if the appropriate personnel do not fix that simple piece.
Communications. Communication failures probably make up the highest number of incidents. Since most financing requires payment for energy, a disruption of communications that results in the loss of production data may have the same impact as a disruption of service. The largest challenge is to maintain a reliable service that provides continuous access to the data. In addition, contracts may require that the owner make access to the data available as part of keeping the warranty fulfillment and performance obligations. Overall, these often-overlooked system components can have considerable impact on the revenue and the risks.
LAYERS OF PROTECTION
Somewhere, fire ants are eating wire insulations. Weeds are growing on a rooftop system. Panels are being stolen. And a delivery truck is about to back into a solar carport. Predicting what could happen is very difficult. The goal is to have suf- ficient expertise and funds available to cover the small things that are not predictable. All of the following layers of protection may not be necessary, and being overprotected may create very limited value. The fundamental component is having a strong team that understands the challenge and can find solutions, quickly.
The first step is selecting the right partners. This process involves asking the right questions and having a dialogue with potential installers, EPC contractors, equipment manufacturers and O&M providers. For example, knowing how many years vendors have been in business provides a level of comfort for developers. Developers can evaluate a company’s solvency by asking to see 3 years of financials. Red flags include a significant amount of debt and an imbalance in cash flow.
Another important factor is the EPC contractor’s strategic partnerships. These partnerships are a reflection of the company. They provide a sense of how the company is viewed and where it sees its weaknesses. Important questions to ask include whether the company is ISO 9001 or ISO 14001 certified. Ask to see copies of safety plans—and read them. The safety plans can give you an indication of how diligent an EPC contractor’s partner is when it comes to risk.
It is important to not be overwhelmed by the sheer number of vendors and product offerings. Healthy competition is necessary to get the best price. However, the benefit in reviewing multiple players must be balanced with the serviceability of fewer relationships. Operations with multiple contractors across multiple inverter and module manufacturers can create hundreds of permutations of problems. It is important to carefully select solid providers that will be able to deliver quality products safely, on budget and on time.
It is also crucial that these vendors and service providers continue to exist to maintain a long-term relationship. For example, when failures occur during construction or after commissioning, the EPC contractors may be the original source of these failures. EPC contractors, especially in this economic climate, can become overextended and potentially go bankrupt.
The most important level of protection for the owner/operator is the EPC provider’s contract. The EPC contractor holds, or wraps, the project risk through construction and potentially through initial operations. In general, an EPC contract should guarantee three things. First, EPC contractors should be able to understand the site, interconnection and equipment conditions to ensure that the system can be interconnected successfully. This means the EPC contractor should take on all the permitting risk and technical obligations to ensure that the design and installation work. Second, the EPC contractor should take on workmanship risk. Third, the contract should contain availability or performance guarantees that provide at least 1–3 years of support.
A performance ratio guarantee ensures a minimum amount of energy delivered versus the amount of irradiation available. These guarantees take into account force majeure events and warranty defects, and they cover much more than workmanship: They are truly an assumption of the risk of the system to perform. Performance ratio guarantees are becoming more prevalent and can last from 1–5 years. Large EPC contractors, however, are not accustomed to guarantees that last for more than 1–2 years. It can be challenging to negotiate these guarantees to assure all parties that their risks are being managed.
Performance ratio guarantees also constitute operations and maintenance agreements. Obviously, maintenance is essential for the system to be fully operational. An implied O&M agreement through a performance ratio guarantee, however, may not be explicit on the level of service or the consequences for a lack of that service.
Procurement agreements are important to reduce risk for the EPC contractor or owner who is wrapping the panel risk. The first step is to insist on flash-test data for every module delivered. The market generally charges in terms of dollars per nameplate watt. Many module manufacturers still offer a +X%/-X% power tolerance. This can mean that a buyer may get less than the nameplate capacity for a large order. Guaranteeing a sum of flash-test data to be equal to or higher than the sum of the nameplate capacity can mitigate this issue. This, however, does not address the challenge of having too much deviation in the module power characteristics. A +/- 5% module tolerance can translate into a deviation up to 10% from the high end to the bottom. The gap is modeled as module-mismatch loss, and it reduces the overall performance. Looking for tight tolerances, hopefully without a negative percentage, can alleviate this. In addition, it is important to review the distribution of production to estimate the amount of mismatch.
Major equipment warranties offer differing degrees of protection against inherent O&M risks.
Modules. The primary potential for production loss is the failure of modules to perform to their specifications. Owners and performance guarantors are not comfortable with the common stair-step power warranties—such as 90% in 10 years, 80% in 20 years—that expose projects to a 10% risk from commissioning through Year 10. Warranties should more closely follow the expected degradation of modules. Solar- World, for example, offers a 25-year linear performance guarantee on all current modules. This guarantees 97% of nominal rated power in Year 1 and a decline of no more than 0.7% per year, guaranteeing 80.2% of nominal rated output in Year 25. This linear performance guarantee increases the risk to the module manufacturer.
In addition, warranties should be protected with a significant warranty reserve or bonding program to ensure that the product will be around when you need to make a claim. According to Benjamin Compton, vice president of PV systems and services at United Solar Ovonic, “With the rapid pace of change in the industry, the biggest obstacle to long-term PV system operations is obsolescence of parts, such as the inability to find PV modules or inverter components after even a few years.”
The form factor, voltage and current characteristics for modules are continually changing. A module purchased today may not be available in 5 years. Therefore, remember that warranty fulfillment may be a check for several hundred dollars rather than replacement product, which will not help your orphaned string. Ordering one to two strings per MW worth of spares modules with the initial large purchase can allay these concerns. Also, a small order later may not have the same terms as the original multi-MW order. The orphaned string problem can also be solved with new products, like dc-to-dc converters that optimize performance at the module or string level. These products add cost but can provide some mitigation in the future.
Inverters. Despite manufacturer’s best efforts and intentions, inverters remain the Achilles’ heel for a solar array. Even if a system has an excellent performance ratio, the inverters need to be on to harvest electricity. While a perfect product is not required, perfect service is. Inverter manufacturers provide a wide variety of services. In general, inverter procurement agreements have been shifting from simply providing expensive power electronics to offering equipment plus service contracts.
Inverter manufacturers understand that developers, investors or utilities do not necessarily want to learn to maintain a new piece of equipment. They just want the service of dc-to-ac conversion to work perfectly. In this case, the service of transforming dc electricity to ac electricity can take the form of 5-, 10-, or even 30-year service and availability contracts. In practice, these long-term performance guarantees and availability contracts are effectively O&M contracts, as the performance targets guaranteed cannot be met without ongoing O&M. These contracts must still be managed carefully, however.
The US market has dozens of inverter manufacturers. It is crucial to work with an organization that is capable of providing repairs and getting the system operational quickly. Even for experienced service personnel, it is challenging to learn to troubleshoot two to three manufacturers’ products, let alone 12. New product lines can be substantially different, and a strong technical expertise on one inverter may not translate to another product from the same manufacturer or from different manufacturers. Therefore, the best mitigation is to have a clear availability guarantee that draws a distinction as to what constitutes availability.
Operations start with comprehensive commissioning, which is not just for installers. Owners, investors, developers and O&M providers need to be proactively involved. The first step to a successful start-up is to have all stakeholders prepared and available. The best place to start is to “yellow-off ” the design drawings with the engineer, constructor and operator—physically walking through the end product with the design drawings and using a highlighter on every line to ensure that what is in the field exactly matches the drawings. This can be time-consuming, but it may be the last time the intentional and the actual are compared.
Commissioning should be a formal process that is explicitly detailed in the EPC contract. The independent engineer should be present as boxes or ditches are closed up to verify that the appropriate wires are in place, as they are likely to remain in place for 30 years. The major components—communications, meteorological station, modules, wiring, inverters and interconnection— should be visually checked for quality and completeness. A variety of proof-of-life tests, including resistance to ground, appropriate fuses and continuity, should be conducted. The tracker components should be checked for accuracy. It is also a good idea to cycle the tracker mechanism dozens of times with a high-speed motor to ensure that all components work together and thus avoid infancy problems. Tracker mechanisms usually have issues due to stress in the mechanical assembly. Therefore, it is essential to see that the structure is aligned within the specifications.
System performance should be checked thoroughly after the installation is mechanically assembled. The EPC contract should make clear that payment for work is conditional upon commissioning tests. These tests should include but not be limited to the following: open circuit voltage, I-V curves for every string in several combiner boxes, operating voltage, operating current and 30-day operating performance tests. These commissioning checks are benchmarks against the next 30 years. It is common to rush through the commissioning stage, to reach financial closure, but it is a bad idea: This is the last point of leverage with the EPC contractor. Commissioning without adequate descriptions, time and available resources leads to immediate and long-term risks to the investment. (For more information, see “PV System Commissioning,” October/November 2009, SolarPro magazine.)
Insurance—the final layer of protection for site operations— involves paying someone else to take risk away from the project. For better or worse, that someone is actually a group of people, typically a combination of underwriters who may not understand the risk and individuals close to the deal. Therefore, it is important to buy insurance where it is needed without overinsuring or paying too much. Do not pay someone else to take on a risk that you can afford to assume. Everyone on an operational team should read the property and general liability insurance policies. They can be dull reading, but they help the team understand the day-today risks they are taking.
The first step is to find a good broker. Many brokers have a variety of skills and expertise in all kinds of insurance. There are brokers skilled in property coverage and those good at renewable energy. However, there are very few that offer both. Brokers are important because of their relations to the insurance underwriting community. The better the underwriters understand the business, the better they will be able to assess the deal. Most brokers approach the same group of underwriters, typically in London or Bermuda. It is critical to explain any project’s specific needs to the broker.
A good insurance program is only worth as much as can be claimed. The process of filing and walking through a claim takes practice. It is essential to have a good broker and trusted advisor when you have the unpleasant task of making a claim. A good insurance program covers catastrophic losses and meets investor requirements. However, investors may require coverage that does not appear to make sense. Investments can have a pool of investors and thresholds set for reasons that are not explicit to solar. For example, European investors typically have set very low deductible thresholds to insure essentially every claim. American investors, however, tend to have a higher threshold for deductibles. The deductible amount should be available in reserve at the project to insure the asset. Generally, on projects greater than 1 MW, a $50,000 deductible seems like a good compromise, balancing claim size with a lower premium. This is a negotiation and should be evaluated on a project basis. These policies should cover all perils, with explicit named exceptions.
Overinsuring is a waste of assets: Replacement value is just that, not a check sent in the total insured amount. The cost of replacing a system might be significantly lower several years from now than it is today. Therefore, it makes sense to insure only to slightly more than estimated costs to replace.
Unfortunately, many large projects are proposed in areas with earthquake, flood or wind activity. It is crucial to evaluate such risks early in the development cycle. Underwriters manage these perils differently, except for one thing: Their policies are expensive. Earthquake coverage is a good example. It is unlikely that an entire portfolio of projects will be affected by a single seismic event. Insuring the total value of every deal is probably not wise. A good broker uses sophisticated modeling techniques to get a good estimation of the probable maximum loss. A company’s risk management program should then evaluate the risk to see if the consequence is tolerable.
Case Study: Fetzer Winery, Hopland, CA
In October 2006, Powerlight and (then) MMA Renewable Ventures installed a 900 kW dc roof-mounted PV plant at Fetzer Winery in Hopland, California, under a power purchase agreement. MuniMae owns the project and dissigno operates it. The plant is spread out on three building roofs with a total of 4,992 Sharp modules covering a surface area of 69,912 square feet. Since the commercial operation date, the system has produced approximately 3.4 GWh of PV energy.
The O&M for the Fetzer plant is fairly standard, except for one unique situation. The array located on top of the building known as the Red Barrel Room (RBR) demonstrates lower-than-expected performance every year. Performance does not drop at one particular time, but slowly decreases. Close inspection of the modules over time reveals very noticeable soiling. Under the air vent, where air blows across the modules rather than up to the sky, a substantial amount of mold grows on both the roof and modules. The glass becomes dull, dark in color and opaque.
The RBR is where the red wine is poured into aging barrels and stored. During the filling process, a winery employee fills the barrels to remove all air. Excess wine spills to the floor, and the barrels are sealed. Because this is a wine storage and aging warehouse, the ambient temperature is kept at 69°–70°F, with high humidity. These two factors create an ideal environment for mildew growth. The ventilation in the RBR draws air in at the base of the building and expels it through a vent on the roof. Mold is transported through the building ventilation system and deposited on the roof and PV modules.
Because the Fetzer plant is covered by an annual performance guarantee, it is up to the O&M provider to determine if it is better to wash the system or pay damages for underperformance. Performance measurements over time indicate when the mold buildup begins affecting production. This drop in performance can be as much as 30%. Once the mold was discovered, annual module cleaning at the RBR was incorporated into the regular maintenance schedule. To determine the best time to clean, dissigno models the losses due to mold. Table 1 shows the difference in production before and after cleaning in March 2010. Washing the modules ensures that the system performs above the required levels.
A simple, noninvasive cleaning process has proven successful. The cleaning crews use soft brushes and pressurized water to wet and wash the modules, and then use squeegees to clear excess water to avoid mineral deposits. The crews do not use chemicals or solvents, only water that comes from the local tap and passes through a filter to remove large dirt particles. The crews, who rely on their prior window-washing experience, also received training from the O&M company. They were taught to avoid damage to the modules by walking on the frames as they clean, to wear soft-soled shoes and to use fall restraints.
A PV system is a financial investment that pays over a very long period of time. The tendency to model a low O&M fee increases profitability. However, this fee needs to have more scrutiny as more long-term information becomes available. These O&M contracts may become net losers for the provider. As this market continues to stabilize, net losers are bad for both providers and for those getting the service. Vendor solvency is essential.
O&M can be a lucrative opportunity for companies with PV experience. Due to warranty and guarantee expectations, it will become increasingly difficult for small-to-medium size companies to act as EPC contractors. At the same time, potentially higher-margin, long-term O&M contracts need to be provided locally. The renewable O&M market is fragmented across the US. There are good-quality companies providing good service, but there is no simple way to tap that experience and get a consistent level of service across a portfolio of assets.
One of the challenges is knowing what O&M costs are in specific markets. The markets in Texas versus California versus North Carolina are dramatically different depending on labor rates, expertise and project volume. Operations are local to the site and need to be priced as such. As operators better understand the requirements, they will be able to increase their margins. However, owners who have a portfolio of projects need a consistent approach to all of their assets. Modelling a good estimation of O&M costs is crucial.
Due to the Federal Investment Tax Credit, it is advantageous to incorporate the O&M costs into the price of the system. These costs are generally limited to 5–10 years. However, prepaying for O&M can make it difficult—if not impossible— to get the services promised. Several large companies offer long-term service contracts but have limited operational resources to monitor, evaluate, prioritize and solve issues. The benefit of prepayment needs to be weighed against the consequence of reduced leverage with the O&M provider. This can be slightly mitigated by defining the expectations and consequences for a lack in performance in the EPC contract.
In 5 to 10 years, many integrators will not exist in the form they do today, potentially creating orphaned asset management responsibility. The market is changing dramatically, and fewer companies will be capable of managing existing contracts and obligations. Being confident that the required work will be done and knowing who will do it is crucial. The assignability of the contract needs to be reviewed to ensure that the qualifications from one vendor are not lost if that vendor assigns the project to another. In reverse, the owner may want to transfer the contract for a better performer. Every contract should include a termination for convenience clause.
Predicting the actual O&M cost dramatically affects profitability if it is incorrect. A set level of scheduled maintenance should meet the stakeholders’ risk tolerance. The level of insurance and projected costs for unscheduled maintenance needs to be considered as part of a comprehensive operations expense. Defining the level of certainty is essential for developing the pro forma project analysis necessary to evaluate potential profitability.
Project owners may be overpaying for O&M support by including too much preventative maintenance. A good evaluation is to weigh the expense of preventative versus reactive project work. Washing the modules is a good example. This tends to be included in most scheduled maintenance, but the timing and frequency of washing is highly dependent on actual weather conditions. The challenge is that long-term investors need confidence in the project manager or project sponsor’s ability to see that this work is accomplished.
Because operations teams benefit from scale and lessons learned from other projects, every developer or integrator does not need to create an operations team. There is very limited upside to this and mostly downside. Operations teams need experience to understand the technical components, contracts and financing to protect everyone’s investment. Also, most of the value of a project is long gone by the time of O&M. The EPC contractor, equipment suppliers, tax equity, traditional equity, landowners, developers, brokers, insurance underwriters and taxes have all taken their piece.
MANAGING THE UNKNOWN
Independent engineers determine a 30-year estimate for operations. This is the number used by financing partners to look at the long-term debt and equity returns. This number is based on a lot of uncertainty, including equipment failure, weather related fatigue and inadequate maintenance, all of which are hard to predict. However, a competent independent engineer and an experienced operations team should provide the expertise needed to identify the risks, probabilities, consequences and solutions.
O&M creates an obligation that exists for decades. Understanding the parameters and contracts for each deal is crucial, because each one has a different level of exposure and costs. The project sponsor needs to seek good advisors to determine the projected costs of scheduled and unscheduled maintenance. Addressing risks begins with selecting the vendors, negotiating contracts, understanding what breaks and how to resolve challenges.
O&M AS A MARKET OPPORTUNITY
The operations and maintenance of large systems is not a mature service industry. Many of the local providers and national equipment manufacturers count service technicians by the tens, not thousands. Guido Tonin, executive vice president of Santerno, explains: “Solar fields, regardless of size, are a specialty niche application today within the larger energy production market. Dedicated expertise matters, so long as O&M providers meet the necessary business viability prerequisites.”
The expanding space and size of the solar industry requires significant development in the O&M area. As solar is growing into new regions and larger system sizes, the infrastructure is lagging behind. For example, it is difficult to find solar panel cleaning companies and O&M providers with adequate insurance and experience. Large EPC providers who are forced to take on a large balance sheet risk are looking locally for a regional presence to solve O&M challenges because they will quickly move onto the next project. This market maturation creates an opportunity for companies with solar installation experience.
This growing market is potentially good for integrators, even those losing work to large construction deals. According to United Solar Ovonic’s Compton: “The larger integrators are having difficulties staffing appropriately and training adequately to keep up with the growing rate of their installations. They will rely more and more heavily on externally subcontracted resources in the near future. Additionally, PV system fleet owners will look to consolidate their O&M services with a single provider.”
A local company can provide operations services to a site near its business. Large constructors, however, cannot effectively staff every site as the scale and diversity of solar plants continues to grow. Fotowatio Renewable Ventures’ Disch thinks there is a market for small-to-medium size O&M providers. “There are benefits to having a local O&M provider, such as lower travel costs and reduced downtime with more rapid response,” she states. “In addition, site security may be nominally improved with a local presence.”
Ultimately, the men and women who actually turn the wrenches are the people who make these PV systems work. These individuals have to navigate contracts, investors and multiple stakeholders with conflicting perspectives, as well as fire ants, hail, rattlesnakes and the occasional ornery security guard. It is important that they understand the risks inherent in O&M, as well as the opportunities for business development and diversification.
Dave Williams / dissigno / San Francisco, CA / dissigno.com