Author Archives: Chris Williams

About Chris Williams

Chris Williams works with HeatSpring developing products and managing online content. Chris is a NABCEP Certified Solar PV Installer and an IGSHPA Accredited Geothermal Installer. He has installed over 300kW of solar PV systems, tens of residential and commercial solar hot water systems and 50 tons of geothermal equipment. You can find him on Linkedin

What Commercial Solar Developers Need to Know about Yieldcos

If you’d like to listen the interview, please scroll to the middle of the article. 

There’s been a huge amount of buzz about yieldcos and their potential impact on financing more commercial solar projects by providing cheaper capital. There have been 8 IPOs in 2013 and 2014 with many more on the horizon.

If you need to brush on details of financing commercial solar, sign up for our free course on Commercial Solar PPAs 101. If you need to learn exactly how to finance commercial solar projects from start to finish, sign up for our 6 week Solar Executive MBA that starts this week.

There have been some basic descriptions of yieldcos in the trade press, but nothing has described what commercial solar installers need to know about this new legal structure or how it could impact their day-to-day business.

This article will try to translate some of the buzz and address the questions that commercial solar installers will likely have. For example:

  • Do yieldcos have a tax appetite?
  • How will yieldcos impact that day-to-day operations of a commercial solar developer or EPC focused on mid-market projects?
  • When should you start doing research on yieldcos?

What is a Yieldco? Why are they useful?

I’ve listed a number of my favorite yieldco articles at the bottom of this article, but this article on SunPower entering the yieldco market does a good job of describing the basics and attractiveness of yieldcos in the solar market.

Yieldcos offer among the lowest costs of equity funding for renewable energy projects for several reasons. Firstly, these companies generate stable and predictable cash flows by selling electricity under power purchase agreements and distribute most of their cash through quarterly dividends. Secondly, the model allows investors to single out the cash flows generated by the power generation assets without giving investors exposure to other aspects of the parent company’s business. Additionally, Yieldco investments are quite liquid, since they trade in the open markets.

Yieldco 101 Target Audience

This article is for three audiences:

  1. Professionals experienced in commercial solar development who don’t have extensive experience in finance.
  2. Commercial solar EPCs who want to start developing their own projects.
  3. Solar professionals who have heard of yieldcos but don’t know what they are or how they could impact their business.


Note: For the most part, I called the person’s name who was answering if you couldn’t tell. In then first question, it is, Tom Konrad, Chris Lord, Keith Cronin.

Here is who I connected with in the interview:

  • Tom Konrad – Tom runs a blog called Alternative Energy Stocks. He also writes for Forbes and is Director of Research at the JPS Green Economy Fund. In the interview, he provides the perspective of the public equity markets.
  • Keith Cronin – Keith runs SunHedge, a solar consulting firm. Keith sold his solar company to SunEdison in 2007. Now he provides consulting to solar developers, EPCs, property owners, and high net worth individuals. In the interview, he provides the perspective of the solar developer.
  • Chris Lord – Chris is a former lawyer with deep banking experience. He is the principal at CapIron Inc., a firm he created to provide advisory and consulting services to customers, owners, developers, utilities, suppliers, installers and distributors covering the full range of value-add in renewable energy and energy efficiency. In the interview, he provides the perspective of the private investor community.

Here’s what you’ll learn

These are somewhat in order, but we did jump around.

YieldCo 101 

  • What are yieldcos? What’s the buzz? Why do people care about this within solar industry?
  • Do yieldcos have a tax appetite? Can they provide tax equity for my project?
  • What is a yieldco, and how does it differ from independent power producers and other entities that own or control energy generation assets?
  • What does the best deal for a yieldco look like? How is it structured? What kind of an off-take? Utility-scale versus DG commercial? What projects are economical?

Yieldcos and Public Equities

  • What are your top yieldco stock picks?  Any to avoid?
  • Why pick these versus traditional stocks?
  • How will interest rate changes impact yieldcos?

Commercial Solar Implications

  • Do commercial solar installers need to spend any time thinking about these? (Less than 5MW)
  • When do commercial solar installers need to pay attention to yieldcos?
  • What should a developer consider before partnering with a yieldco or selecting one as a buyer?
  • How, specifically, would working with a yieldco change a deal? (Not flexible in structure or due diligence)
  • What kind of rates are yieldcos looking for in transactions for cash equity? For debt?
  • How to think about starting a yieldco versus owning your own assets.

Conclusions / Parting Thoughts

  • Will yieldcos revolutionize or otherwise open up the world of renewable energy finance, and particularly solar?

Here are the articles I’ve found to be most useful. If you’re completely unfamiliar with the term, these would be useful to read through before listening to the interview.

Posted in Solar Photovoltaics | Tagged , , , , , | 1 Comment

How Massachusett’s Alternative Energy Credit Prices Will Impact Heat Pump and Biomass Operating Costs


We passed the Massachusetts “Clean Heat Bill” in July 2014. The final bill number is S. 2214. 

The bill created a production-based incentive, similar to solar PV renewable energy credits (SRECs), for renewable thermal technologies including solar thermal, air source heat pumps, ground source heat pumps, solid biomass, and biogas.

In this article, I’ll explain how Alternative Energy Credit (AEC) prices will impact the cost to deliver heat from biomass and air source and ground source heat pumps. An AEC works the same as a REC, but the “A” stands for “Alternative” rather than “Renewable.” While the bill applies to more technologies than heat pumps and biomass, I’ll focus on technologies that can be used as primary heating sources for Massachusetts properties to provide some perspective for the building and HVAC industries.

I’ll discuss the impact of a $20/MWth payment on the generating cost for air source heat pumps, ground source heat pumps, and advanced biomass boilers. The current market price for APS AECs is trading around $29/MWth, close to the ACP payment, so $20/MWth is a conservative market price for these AECs.

The article will explain that potential AEC prices will reduce ASHP operating costs by 13%, will reduce GSHP operating costs by 33%, and will reduce biomass operating costs by 30%.

For this article, I’m only going to focus on the impact of operating costs for these technologies. I won’t focus on how these operating costs savings will impact the returns of specific projects. The reason for this is that it would require too much research and I’d have to control for to many “what if” scenarios. In order to address returns on a project, I would need to include installation costs and then compare these installed and operating costs to a competing technology to determine how it would impact returns and savings.

My assumption is that by providing specific operating cost information you can then apply these to your project-specific installed costs.

If you need some technical background information on how air source heat pumps, ground source heat pumps, solar thermal or biomass work, sign up for our in-depth free course on the subject: High Performance Building and HVAC 101

Background on the Bill

S.2214 created a production-based incentive for “useful thermal energy” that provides heating and cooling in situations where fossil fuels would have otherwise been the source of energy.

Here is how the bill defines useful thermal energy:

“Useful thermal energy”, energy in the form of direct heat, steam, hot water or other thermal form that is used in production and beneficial measures for heating, cooling, humidity control, process use or other valid thermal end use energy requirements and for which fuel or electricity would otherwise be consumed.”

One of the most impressive aspects of the bill is that it had broad support from many parties including private industry, environmental groups, renewable energy groups, electric utilities, and the Patrick administration.

What made the bill work well is that the existing Alternative Portfolio Standard (APS) market was under-supplied. The existing APS was created in 2008 as part of the Green Communities Act. However, the applicable technologies in the APS have never fully developed. The result of the APS under-supply is that the utility load serving entities were paying alternative compliance payments (ACP) because the AECs didn’t exist. The total ACP payments totaled around $12MM per year.

Assumptions for the Cost to Deliver 1 MMBtu For Each Energy Source

To discuss this, we first need to make a number of assumptions about each technology source.

Air Source Heat Pumps Operating Costs

  • Electricity Cost – $0.15/kWh
  • Operating COP – 2. Many will say that this is low. However, for my analysis, I’m assuming that air source heat pumps are the primary and only heating system for the entire load. This makes the analysis easier on my end, but would mean a lower COP over a long period of time over a large data set of systems. If you have any specific data to the contrary, please put it in the comments.

Ground Source Heat Pumps Operating Costs

  • Electricity Cost. $0.15/kWh
  • Operating COP – 3.5. With this, many in the ground source industry will say that this is too low and that they see COPs in the 4s and 5s. My response to this is that I’ve never seen any real time data over a large number of years, with a large sample size, that can support this claim. Read more about real time geothermal monitoring data. An operating COP of 3.5 is much more conservative.

Biomass Boilers

  • Cost per ton delivered: $225. This is based on conversations I’ve had with a few suppliers.
  • BTUs per ton. We will assume that each ton of pellets can produce 16 million BTUs.
  • Operating efficiency of boiler and distribution system is 80%.

Based on these assumptions, here’s the cost to deliver 1 MMBtu for these systems.


Here are the specific numbers:

  • ASHP – $21. 98
  • GSHP – $12.56
  • Biomass – $17.57

In case you’re curious, here’s the calculation:


For each MMBtu, what percentage came from a renewable source?

  • In the case of air source heat pumps, how much heat was extracted from the air?
  • For ground source heat pumps, how much heat was extracted from the ground?
  • For biomass, how much of the heat delivered came from biomass?

Based on our assumptions of operating efficiency for each technology, here are the answers:

  • ASHP – 500,000 BTUs of 1 MMBtu will come from the air
  • GSHP – 714, 285 BTUs of 1 MMBtu delivered will be extracted from the ground.
  • Biomass – 900,000 of 1 MMBtu will come from biomass.

How many MWth were produced from a renewable source?

AECs are minted on a per MWth basis. Thus, we need to convert MMBtu to MWth.

The conversion to go from MMBtu to MWth is multiplying MMBTU by 0.293. In case you’re curious, the conversion factor to go the other way, from MWth to MMBtu, is multiplying MWth by 3.412.

Here’s how much each type of technology, given our assumptions, will harvest from a renewable resource per 1 MMBtu delivered to a conditioned space.

  • ASHP – 0.146 MWth
  • GSHP – 0.20 MWth
  • Biomass -  0.26 MWth

If we assume that the AEC prices are $20 per MWth, here’s the value of that production per MMBtu delivered.

  • ASHP: $2.93
  • GSHP: $4.19
  • Biomass: $5.27

MMBtu Cost vs. AEC Value vs. New MMBtu Cost

To make this interesting, let’s compare the existing MMBtu delivery cost, the value of AECs per MMBtu with the new law, and the new cost to deliver MMBtu after considering the AEC prices.


 Here is the information on this graph with specific numbers:


What this graph shows is that, given our assumptions about operating efficiency are correct, the AEC prices will reduce ASHP operating costs by 13%, it will reduce GSHP operating costs by 33% and it will reduce biomass operating costs by 30%.

How would this impact normal home economics? Let’s assume we have a house with a 100 MMBtu load. 

See the graph for what the numbers what would look for a 100 MMBtu load.


You’ll notice that the AEC prices, while they do decrease the per MMBtu cost by between 13% and 30% is substantial, they don’t add up to a large amount in cash.

Here is what the AEC payments would be for each system if the system delivered 100 MBtu in a heating season, given all our assumptions about AEC price, operating efficiency of equipment, and how much it ran.

  • ASHP – $293
  • GSHP – $418
  • Biomass – $527

You’ll notice that these payments are not huge. Given that “revenue grade metering” does not come standard on any of this equipment, this could be an issue for smaller systems.

Heat Metering

The legislation clearly states the metering requirements. You can see the language from the bill at the top of the below slide.


Systems using biomass boilers or ground source heat pumps can be metered effectively. ASTM is currently working on a heat metering standard that should be completed by 2015. However, there are no known methods for providing utility-grade metering for biomass furnaces or air source heat pumps. It’s extremely difficult to measure heat transfer through air.

This is further compounded by the amounts of money that are being considered. For small residential systems, the cost of metering systems, even if a standard exists, would likely outweigh the increased revenue of those systems.

The DOER is currently in the process of creating regulations and a key aspect that they are considering is metering guidelines and how to distinguish between small and large systems.

Upfront Minting – Getting XX years of AEC Payments in Year 1

During our renewable thermal stakeholder metering, one of the things that the DOER expressed interest in is “upfront minting,” which would mean that a property owner would get the credit for many years of AEC payments upfront. The amount of payments would be based on software projections. If the systems had metering and underperformed, there would be some sort of under-performance penalty in future years. There are many “what if” scenarios for upfront minting that the DOER is still trying to figure out.

Upfront minting would have the benefit of decreasing installed costs in year one, something that property owners are extremely sensitive to. Here’s how much 5 and 10 years of AECs could be worth in our simplified example.

5 Years of AECs

  • ASHP – $1,465
  • GSHP – $2,090
  • Biomass – $2,635

10 Years of AECs

  • ASHP – $2,930
  • GSHP – $4,180
  • Biomass – $5,270

Another issue raised by upfront AECs is who will cover the spread. If a biomass system is getting paid for 10 years of AECs before those AECs have actually been created, where is the money coming from?

An answer doesn’t exist for this question yet, but it will be important to consider.

Heat Pumps as “Producing” Energy

One of the key aspects of the law that always brings up an interesting conversion is this line of thinking: “Heat pumps don’t produce energy; they’re energy efficiency. They just move energy.”

My response to that is twofold.

First, solar PV does not produce energy; it just moves it. However, we consider solar PV to be a production resource. It moves energy from sunlight into something that we can use in the form of AC electricity. Also, solar PV isn’t very efficient at all. It only converts about 15% of 20% of the sun’s energy into useful energy. You could argue heat pumps do the same thing, they move energy from a non-useful to a useful form. However, heat pumps are actually much more efficient than solar PV.

Second, if I had a air source heat pump that delivered 10 MMBtu to a conditioned space with an average annual COP of 2, it means 5 MMBtu came from the outside air. Aren’t the BTUs in that air renewable? Obviously they are.

Applications in the Market

Another question to think about as the regulations for this law are created and go into effect is how and where the new law will impact the existing heating market. There are two places to look at: the residential HVAC retrofit market and low energy use building market.

Residential HVAC Retrofits

Massachusetts is a retrofit market. This means that the large benefit of these incentives will be to spur investment in these technologies for existing homes. However, metering on a residential project, assuming AECs cannot be minted upfront or a simpler method can’t be created, could be cost-prohibitive.

If the DOER can figure out how to minimize metering costs and pay for AECs upfront on smaller projects, the residential market will benefit enormously.

For the commercial and industrial markets, project costs relative to metering costs will be so large that metering won’t be an issue. Also, the AEC value for larger projects will be much more substantial.

Low Energy Use Buildings

While low energy use buildings are a growing trend, they’re not a large enough segment to actually impact the market. However, within these buildings, air source heat pumps tend to be the main source of heat pump simply because the space heating loads are so low. In these instances, almost by definition, they wouldn’t create many AECs simply because they don’t need much heat.

Further Learning

Here are a few resources if you’d like to learn more about the basics of these technologies, the new Massachusetts renewable thermal law, or existing renewable thermal incentives in Massachusetts.

  1. Free Course: High Performance Building and HVAC 101. This is an in-depth free course on high performance building, air source and ground source heat pumps, and biomass HVAC systems
  2. Massachusetts Clean Heat Bill
  3. Existing Massachusetts Renewable Heating and Cooling Incentives


Posted in Uncategorized | 2 Comments

Community Solar 101

There’s been a lot of buzz about community solar lately. While the amount of investment in community solar projects is a fraction of the investment of the entire solar industry, there are a few leaders who recognize the incredible potential of community solar.

Over the next few months, we’ll be publishing a series about community solar. This is the first article in that series. If you’d like to learn more about the subject, sign up for our free Commercial Solar PPA 101 course.

In this article, I’ll address the following questions.

1. What is community solar?

2. Why is community solar important for the solar industry?

3. What are the regulations that make community solar possible?

4. Who are the key parties involved?

5. Are the different structures that can be used to develop community solar projects?

Background – Why are we writing this article?

To give you a sense for how small community solar is at the moment, SEIA reports that there “at least” 52 “shared renewables energy projects” in the US. While it’s likely the amount is higher, even if it’s 10 times more, that’s only 520 projects.

We decided to create this article series simply because of the volume of questions we’ve been receiving about this topic in three places.

  1. A section in our commercial solar PPA 101 free course dedicated to community solar has been getting a lot of questions.
  2. Professionals continue to join our LinkedIn group “Best Practices for Financing Mid-Marketing Solar Projects” and ask for help on community solar topics.
  3. Lastly, in our Solar Executive MBA, where students are focused on learning how to develop and finance commercial solar projects from start to finish, we’re getting more and more questions about community solar.

This article will outline some of the basic key points around community solar. The goal with this first article is to go much deeper than most of the trade press can go and provide practical advice on the subject. While some of the content will be original, a lot of what I’ll focus on will be curation and organization of the existing material on the Internet that will make it easier for you to do research on your own time.

We’ll start with the basics and get more and more detailed.

There are so many questions to be answered that it’s difficult to determine where to start, so I asked our “Best Practices for Financing Commercial Solar Projects” Linkedin group, focused on financing mid-market solar projects any questions about community solar, to tell me what they wanted to learn. Here’s what they asked:

  1. How much should a developer expect to pay to lease the land/rooftop space for the array?
  2. Are lease payments based on $ per acre (sq. ft.)? or $ per MW? or $ per MWh?
  3. Are there typical escalators? 1% a year?
  4. Who takes responsibility for paying the taxes? Landowner? Developer?
  5. Are typical lease terms 20 or 25 years? Do they typically have options for renewal?
  6. What sort of developer fee is reasonable given the significant additional overhead associated with managing a community array?
  7. Do the community solar customers really own a piece of the system? Or do they only own the rights to the energy and environmental attributes of the production from their piece of the system?
  8. Is the community solar LLC able to depreciate the value of the solar array?
  9. Who gets the federal tax credit? The community solar owners on a pro-rata basis? or the community solar LLC?
  10. What are the advantages / disadvantages of having a large anchor tenant / off-taker?
  11. Are there any rules of thumb on the cost of sales? How much should a developer budget in cost per watt for sales/marketing/admininstration/legal expenses to close a customer?
  12. Do you have any legal agreement templates to review for several agreements needed between the various parties?

These are all excellent questions. They’re just hard to answer without a significant amount of work. Also, given that such a small number of installations have been completed, it’s not clear that there are concrete answers for any of these questions.

By the way, if you’d like to connect with other professionals working on financing community solar projects, please join our Linkedin group on best practices for financing community solar projects.

For this article, I’m going to assume that most solar professionals have heard of community solar and know that it’s something about having one large array that credits a larger number of residential customers but have not done much more reading than that. If you have a much deeper understanding of community solar than this, this article will be dull to you. If this is exactly what your understanding is, it should be a better fit.

What is community solar?

In the most basic form, community solar means that there is a single solar asset that is producing power. The power that is produced can be owned or purchased by multiple parties that are not sited at the exact location of the array.

In many ways, community solar is similar to community wind, although community wind has been far more successful. The wind industry, unlike the solar industry, first developed as a utility-scale energy provider and has slowly been working towards smaller and smaller projects. Two of the largest community wind developers, National Wind and OwnEnergy, have developed a combined capacity of 5,000 MW.

How does community solar work?

  1. There is a single large solar array that is installed.
  2. Similar to financing any commercial solar project, it can be structured as a power purchase agreement, where the homeowner is simply purchasing a specific amount of power, or an ownership model, where the members technically own a certain number of modules of the array and receive the production of those modules.

Why is community solar important for the solar industry?

Community solar is important for a number of reasons.

  1. Increase the addressable market size by 2X to 3X overnight.
  2. Instantly lower the customer acquisition costs for a residential solar installer that is already generating and selling roof mounted solar projects.
  3. Lower investor risk, in theory.

Let’s dig deeper into each of those.

First, community solar drastically increases the number of people that can buy solar. Increasing the number of potential customers means that the solar industry simply has more room to grow, more equipment can be installed, more people employed.

Faze1 screened all of the 1.2 million single family homes in Massachusetts and found that only 26% of them have suitable roofs for solar. It is true that some of them could install a small ground mounted system in their yard. However, it’s reasonable to assume that this would be a small percentage. Community solar provides solar access to the rest of the homeowners who don’t have proper roofs.

It also provides solar access to renters who cannot buy solar because they don’t own their home. In the existing solar model, solar makes the most sense when you own the building where the solar is being installed. Because community solar can change ownership quickly, renters will have access to solar.

The benefit of having more potential solar customers is clear. By increasing the potential market size of solar, it allows there to be a larger target market, more customers, more companies, more solar workers, and more cash flowing.

Second, community solar instantly lowers the customer acquisition costs for a residential solar contractor that is already selling roof mounted solar projects.

Reducing soft costs, and specifically customer acquisition costs, has been a focus of the solar industry for the past 18 to 24 months. Allowing community solar development would solve this problem. In fact, a recent solar bill in Massachusetts would have eliminated community solar potential because it removed virtual net metering. 

I’d argue that the most important reason for more community solar development is the ability to decrease customer acquisition costs for existing roof-mounted solar providers by at least 50% to 70% overnight. Let me explain why.

Let’s look at the sales funnel of a typical solar customer. This data was provided by Faze1. They have done extensive research on optimizing solar marketing profitability using better consumer data.

Let’s assume this is what a typical sales funnel looks like. Yes, these are averages and over-simplified, but they will illustrate my point. You could simply plug in your business’s number to get a better idea.

  • Marketing spend: $10,000
  • Leads Generated: 300
  • 30 – Qualified lead. Those that are willing and able to go solar.
  • 120 – Willing and not able lead. Willing means that they are interested in solar, have good credit, etc. They could purchase cash, use a solar loan, or buy cash. Not able means that the existing site is not suitable for solar.

This is anecdotal evidence from most of the contractors I’ve spoken with. But it’s clear that in order to find customers who can go solar and have acceptable roof space requires attracting and talking with 3 to 5 times more potential customers who want solar but don’t have the roof space. By being able to sell that customer a 20-year PPA for their home’s power or a 5kW share of a 1MW community solar facility, you can generate more revenue for the same marketing spend and number of salespeople.

If selling a community solar share was possible alongside roof-mounted solar, then the same $10,000 investment in marketing would yield 150 qualified leads instead of just 30.

Here are the numbers from my simple example.

  • Cost per qualified lead without community solar; $333 ($10,000 divided by 30)
  • Cost per qualified lead with community solar: $66 ($10,000 divided by 150)

Third, community solar has lower investor risk, in theory.

With community solar, the risk of default is lower than a PPA with a traditional roof mounted system. The reason for this is simple: In the case of non-payment, the community solar provider can instantly find another customer and change who is being credited for the power. In the case of non-payment for a roof-mounted project, power can be shut off from the solar provider, but there is no easy way to recoup the value of the solar array.

However, I’d suggest that, while this theoretical reduction in risk is true in the long term, community solar is perhaps a little more risky in the short term from an investor’s perspective simply because it is new. Potential risk will be affected by regulations, policy, and execution.

Key Parties

Here are the key parties in a community solar project.

  • Community Solar Service Provider. The community solar provider is responsible for setting up the SPE, gathering members, changing members, and handling billing.
  • Special Purpose Entity (SPE). The SPE is the specific legal framework that is set up to finance the project. It’s typically a LLC and it’s set up to own and operate the project.
  • Subscribers or Members. The subscribers or members are the “off-takers” for the project. They are buying the power. If they are “members” and invest in the project, they contribute their own money to buy a part of the project.
  • Host. The host is simply the location where the physical array exists.
  • Utility. The utility is responsible for distributing the power and billing credit. In the case of a utility-sponsored model, they are also buying the power and then distributing it to their members.
  • Developer. The developer does the engineering, procurement, and construction work and sets up the PPA.
  • Installer. The installer is responsible for building the project.
  • Investor. The investor is the individual or entity that is financing the project and monetizing the tax credits if the community solar project is financed with a PPA.


There are three types of regulations that allow for community solar development:

  1. Group Billing Standards. Group bill is often compared with how master metering arrangements can be set up in real estate transactions. A landlord receives a single bill for the entire building. The landlord then determines how to split that bill up between all of the tenants. Using group bill in the content of solar works the exact same way, except all of the “tenants” don’t need to live in the same building. What happens is that the utility creates a group of members who want to be billed together. The utility produces a bill that describes all of the members’ electric usage and charges. Second, the output from the solar array is netted against the group bill. In this way, a number of residential homes can receive credit from a single facility. In this structure, there must be a single utility representative that deals with disputes and billing. This is the structure most commonly used in Vermont. Read about community solar lessons learned in Vermont.
  2. Virtual Net Metering. Virtual net metering allows net metering credits generated by a facility to offset loads at multiple retail electric accounts within a utility’s service territory. Under virtual net metering, credits appear on a customer’s bill as they would under a traditional net metered project.
  3. Joint Ownership. Laws and regulations that allow joint ownership allow many individuals to invest and own a certain percentage of a larger solar array. They are then are entitled to the power produced by that array.

Available States

Technically, there is some form of virtual net metering in 11 states according to DSIRE:


However, only 4 states have effective policies that are actually spurring investment. Those are Massachusetts, Vermont, Colorado, and Minnesota.

This is a development map from Clean Energy Collective. You can see they only develop community solar projects in 4 states.

Screen shot 2014-10-09 at 8.54.07 AM

Why is there a difference between laws on the books and development?

Laws on the books and effective regulations are separate animals. I ran into this when trying to help a friend develop a small community solar facility in Maine, a state that does allow joint ownership and virtual net metering, technically. What I ended up having to do is work through a loophole to make the project the work. However, this loophole is only something that’s possible with close friends or family situations and made it obvious why community solar on a commercial basis in Maine is impossible. Read more about a step-by-step guide to a 14.25kW community solar project in Maine. 

 Ways to Structure Community Solar Programs

In my research, I have found that there are three main ways to structure a community solar facility. All of these images are courtesy of a NREL report on community solar. 

1. Utility-sponsored model

Under the utility-sponsored model the utility itself owns or operates the solar array. Ratepayers of this utility are then allowed to voluntarily chose to receive power produced by solar.

Here is how flows of capital work in the utility-sponsored model.

Screen shot 2014-10-09 at 8.44.56 AM

2. SPE. In special purpose entity formation, a group individual investors join a business enterprise to develop a community solar array.

Here is how this model is structured.

Screen shot 2014-10-09 at 8.45.05 AM

3. Non Project Buyback structure. Through a non-profit entity, donors contribute to purchase a community installation that is eventually owned by a charitable organization.

Screen shot 2014-10-09 at 8.45.16 AM

This is the first article in a series of articles and interviews that we’ll do on community solar. If you have a question about any of the content, please leave it in the comment section.

Additional Recommended Reading

If you’d like to learn more about the topics in this article, I highly recommend these resources.


Posted in Financing | Tagged , , , | Leave a comment

How to Handle Unknown Risk to Increase Solar Project Success

Known Knowns PNGImage: Universe of Issues, Risks, and Challenges

This is a guest article from Chris Lord, Managing Director at CapIron, Inc. He’s a former lawyer with extensive banking experience who now consults with solar developers and investors. I’ve never met anyone else who can, seemingly, answer any financial or legal questions about financing commercial solar projects.

In the article, Chris shares some of his experiences about how to understand and mitigate the risks that you don’t know exist in commercial solar development. Unknown unknown risks are extremely important to understand because they can have large negative impacts on profits and relationships with investors and clients. These risks are especially important for firms that are experienced in solar but new to financing larger commercial solar projects.

I found this article extremely interesting and if your work revolves around selling or financing commercial solar projects, I’m sure you’ll love it. If you have questions about the article, please leave a comment. If you’d like to connect with other professionals focusing on best practices for financing commercial solar projects, join our LinkedIn group on Best Practices for Financing Mid-Market Solar Projects.

Chris Lord also teaches our 6-week Solar Executive MBA that starts on Monday, September 15th. In the course, you’ll work a commercial solar deal from start to finish with expert guidance. The course includes financial models, legal contracts, and development tools that are indispensable.

Enter Chris Lord

Not long ago, I spoke with an experienced developer who told me about a small utility-scale project undertaken by a team within his company. Although experienced with distributed generation projects, the team and its leader had never developed a third party financed, utility-scale project. They knew that they had to learn more about the technical and procedural requirements for interconnection with the local utility and delivery of the solar power to the grid. Over the course of development, the project hit many roadblocks and challenges before finally arriving successfully at COD. Throughout the process, the team modeled the project early and often, generally showing a tight but acceptable profit margin for the project. At COD, the company collected its profit and moved on. Less than a year later, the third party investor in the project made a call on the developer’s tax indemnity required as part of the close. It turned out – to the utter surprise of the project manager and his team – that they had incorrectly assumed the federal ITC would apply the interconnection costs paid to the local utility for equipment on the utility’s side of the transformer. The error – when finally caught – cost the company more than its small profit margin on the project and constrained the company’s cash flow.

This articles focuses on the most dangerous and difficult threat to successful project development: the risks, issues, and challenges that you don’t know that you don’t know. These “unknown unknowns” are not the items that you know you don’t know. When you know you don’t know enough about a risk, issue, or challenge, you can remedy that ignorance by focusing on the problem and calling on experts – colleagues, advisors, consultants or lawyers – to help you learn what you must learn to overcome, hedge, or eliminate it. In the example above, the team knew it had to learn more about the technical and procedural requirements for interconnection with the local utility, and they did so successfully. What the team did not know was that it did not know enough about the ITC’s definition of “eligible equipment” and its application to their project.

Understanding the Challenge of Unknown Unknowns

Developers by nature have to be optimistic and confident souls, if they are to make their way through the minefield of project development. Without that optimism and confidence, a developer would never get started on the daunting task of taking a green field site from start to finish. In fact, the persistence that everyone tends to think of as the critical ingredient in developer success is actually just a manifestation of optimism and confidence.

Known Knowns (PNG)

But as life shows us, our greatest strengths are also our greatest weaknesses. That very same optimism and confidence necessary for successful project development often blinds a developer to the biggest risks of all. These are the risks – that through optimism, confidence, and ignorance – are simply not on the developer’s radar screen. These are not the known or expected risks. A successful developer manages a known risk by minimizing and staging investments of time and money until more about the risk is known or its threat neutralized. There are a lot of surprises in the life of a development project, and, because developers are an optimistic lot, it is rare that these surprises add to a project’s upside. More often than not, these “upside” events were already incorporated into project economics as “good to average assumptions.”

So what really can kill projects are the unknowns and the unexpecteds. We will just call them the “unknown unknowns.” These items consist of issues, events, or results that a developer does not even know that he does not know. And while a wealth of experience and education can reduce the potential unknown unknowns, they are always there. Nassim Nicholas Taleb (author of The Black Swan and several other books on risk) and many other investors specialize in investment strategies designed to capitalize on unexpected and dramatic events, such as the mortgage meltdown crisis of 2008. These strategies involve multiple small bets on a wide variety of extreme outcomes. But a project developer is betting on not having unknown unknowns occur, and that is a lot harder to do.

Tackling the Problem of Unknown Unknowns

The image above illustrates the problem. If we begin with the blue box, then that is the complete universe of all issues, risks or challenges. At the very center of the box is the yellow circle that illustrates what we know (sometimes called the “known knowns”). These are the items that, through education and experience, we know how to handle and are comfortable wrestling with them. The orange cloud surrounding the yellow circle represents the items that we know we don’t know. Within this nebulous cloud are the issues, risks, and challenges that we know just enough about to know we must anticipate and manage them, but we don’t know enough to define them and consider the solutions, hedges, or alternatives. In other words, we know that we can expect the item to arise, and that to manage that item we must either educate ourselves, find an expert to manage it for us, or some combination of the two. For example, most developers know that they must consider whether a project will be subject to property tax over the course of its existence. Property taxes are a set of arcane rules that vary not just from state to state but also from county to county. Moreover, solar PV projects may be characterized and taxed as real property in some jurisdictions, but they may also be taxed as personal property in jurisdictions that make the distinction. In this case, when a developer begins a new project in a new state or county, he or she knows to consult local counsel early – before even meeting with local taxing authorities to discuss abatements or PILOT agreements.

Known Knowns PNG

Image: Universe of Issues, Risks, and Challenges

Specific Actions to Address Unknown Unknowns

So, turning back to our unknown unknowns, how does a developer guard against something that by its very nature is unknown and unexpected? Not easily, of course. But a couple of options come to mind. The key to all of these options is to work on expanding the known knowns and the unknown knowns. If you look at the illustration above, we are talking about expanding our knowledge and leveraging the experience of others to make the yellow circle as large as possible and grow the orange cloud outwards as well. In effect, we want to shrink the blue portion of the box – the unknown unkowns – by expanding the circle and cloud. Of course, we can never eliminate the blue, and should not imagine that is where our efforts should focus, but the faster we can grow the yellow circle and orange cloud, the better hedged against the unknown unknowns we will be.

Continue reading

Posted in Financing, Solar Photovoltaics | Tagged , , | Leave a comment

Why Performance and Not Price Is the Most Important Factor In Finding an Investor or Buyer for Your Solar Projects


Question or comments? If you’re a solar developer or investor and have a story to share that relates to this article or a question about the content, please leave it in the comment section below the article. 

This is a guest article by Chris Lord of CapIron Inc. Chris also teaches our Solar Executive MBA. The next Solar Executive MBA session starts on September 15th. In the course, students will work a commercial solar project from start to finish with expert guidance from Chris along the way. The class is capped so to provide maximum student attention, but there are a limited number of discounted seats. You can get your $500 discounted seat here.

In the Solar Executive MBA, one of the most common topics students have questions about is about identifying, screening, and closing investors or buyers of their solar projects.

Most commonly, students focus exclusively on getting investors who pay the highest price per watt for their projects. In the article on the three keys to defining bankability, we discussed why this is not the best strategy. The investor actually wants the best returns on a project. The best returns means the project is economically strong and reliable.

From the developers’ perspective, there is risk in selecting the right investor. This article will address why it’s critical to address the competence of investors and how developers can screen investors to find the best ones.

Enter Chris Lord from CapIron, Inc

In today’s highly competitive solar PV market, project developers looking for an investor or purchaser for their projects tend to focus almost exclusively on finding those with the lowest project return requirement or willing to pay the highest price for a project.

But is this the best measure to use when locking in your solar project upside?

This article examines the importance of purchaser performance in selecting a project purchaser and outlines ways to collect data that will enable you to assess purchaser performance.

Here is an example of how a developer lost a lot of value in a very short time by ignoring the importance of performance or execution risk when selecting a purchaser for his solar PV projects.

The developer had a mid-sized distributed generation project for sale, largely shovel-ready. The developer asked outside consultants to conduct an auction process among a select group of purchasers. With the bids in, the results were arranged in a matrix to show dollar price against execution risk.

In the matrix, shown below, execution risk was estimated based on a variety of due diligence and market intelligence assessments. The highest execution risk was assigned a ten, and the lowest execution risk assigned a one.


One of the parties added late in the process by the developer offered the highest price at $3.18 a kW, almost $0.35 a kW higher than the average of the other six bidders, and $0.26 a kW higher than the next highest bidder. Based on market experience, the consultants interpreted that as a strong sign that rumors of financial distress at the high bidder were true. The prospective high bidder was desperately trying to bolster a weak pipeline in order to attract a badly-needed infusion of capital.

The consultants recommended a bidder offering a price of $2.85 a kW bolstered by the lowest likely execution risk. Focused solely on price, the developer ignored the recommendation and proceeded with the highest bidder.

After thirty days of intense negotiation on an LOI, and days before execution of the LOI, the purchaser’s parent filed for bankruptcy and the purchaser followed suit. Worse yet, when the developer turned back to the other bidders in an effort to salvage value, he found that they knew of his predicament and were inflexible on terms and soft on their original price bids. Ultimately, the developer settled for $2.82 a kW, but this did not account for the lost legal fees and time spent negotiating a deal that never closed.

1.    Pricing vs. Performance

a.     Why the focus on Pricing?

It is not surprising that project developers zero in on price when selecting a project purchaser. Particularly for small and mid-size developers, finding every possible dollar on the sale price is critical to covering the economic uncertainties inherent in a project’s development and construction phases and generating enough capital to fund continued growth.

The overriding problem facing developers is that there is a complete and natural disconnect between project costs (development and construction) on one side, and the valuation that an investor or purchaser might place on the project.

In the real world, purchasers look solely to the net cash and tax benefits that a project is expected to generate over the 15 to 25 years of its life. By discounting those net cash and tax benefits back to the present using their target return, a purchaser arrives at a price that he or she is willing to pay today for the project and related benefits.

For the capital costs of developing a project, the investor or purchaser is completely indifferent. If a developer spent more than the purchase price, then the developer will lose money. Any amount over the developer’s costs is how the developer generates a return on the development capital invested in the project.  Either way, it has no impact on the value of a project to investors or purchasers.

This sounds simple enough but, given that most developers must find an investor or project purchaser before construction begins, and – worse yet – the actual costs of development and construction may not be known at this point, developers naturally steer to the highest price offered by a purchaser because there appears to be no downside. A higher price gives the sense of security – more margin to cover development and construction unknowns – and, should costs come in at or below projections, more profit to fund future growth.

b.    What’s the downside?

By focusing solely or primarily on price, developers overlook other critical factors including investor or purchaser performance that can dramatically and sometimes adversely impact price. Sometimes this risk is characterized as “execution” risk. Whatever we call it, we are talking about the likelihood and cost of actually closing the specific, targeted transaction with a particular investor or purchaser on terms and conditions (including price) reasonably close to those the parties originally expected when they executed an LOI or otherwise first “shook hands” on the deal.

Performance is important because the ability of an investor or purchaser to follow through and close a transaction in a timely and cost-effective manner can have a bigger impact on a developer’s realized value than the promise of an incrementally higher purchase price from an investor or purchaser who fails to close.

In any financing, there is always a risk that a closing fails. There are at least three main classes of these types of risk: market risk, developer risk, and investor risk.

Market Risk

Market risk is the risk that arises from adverse changes in general market conditions. An example of a market failure, well known to most veterans of solar development, occurred in 2008 with Lehman’s collapse that fall and the onset of the Great Recession. Most project purchasers suddenly lost their tax appetite. Almost all major banks took economic hits to income that saddled them with substantial losses, wiping out the very profits that they were counting on to create their tax appetite. As a consequence, there was very little tax appetite among investors nationwide for the balance of 2008 and much of the first half of 2009. Even when investors did return to the market, tax-drive transaction volume in 2008 was substantially below pre-Lehman projections. In fact, Congress created the Treasury’s Cash Grant program in lieu of the ITC precisely to address that issue.

Continue reading

Posted in Featured Designs, Products, and Suppliers, Solar Photovoltaics | Tagged , , | 2 Comments

AC Coupling – How to Cost Effectively Add Battery Back-up to Existing Grid-Tied Solar PV systems

This is a guest article by Chris LaForge.

Chris is teaching an in-depth 6-week technical training on designing battery based solar PV systems that starts in September. You can read the full description and get a limited-time discount here. If you need to learn how to design, quote, and commission a battery based solar PV array, this is the best course for you.

In the past three years, three trends have converged to create higher demand for battery-based solar arrays: battery prices are declining, the penetration of grid-tied systems is exploding, and homeowners are becoming more interested in backup power.

Retrofitting existing solar PV arrays to include batteries is becoming an opportunity for added revenue for contractors.

Enter Chris LaForge –

AC Coupling

Since the advent of high-voltage battery free (HVBF or grid-direct) solar electric systems, some clients have been frustrated by not being able to use their systems during power outages. The re-work necessary to move to a grid-intertied system with battery back up is costly (GTBB or DC coupled system), inefficient, and, in some cases, unworkable.

Ac coupling can be used in both utility-intertied systems and in off-grid applications. This article will discuss the utility-intertied aspects of AC coupling.

With the advent of AC coupling as a means to introduce battery back-up to an existing HVBF system, an efficient and more workable solution has come to the fore.

AC-coupled systems use the HVBF system while adding a battery-based inverter that works in tandem with the HVBF inverter. It maintains the efficient operation of the PV system while the utility is available and then allows for its operation during power outages by having the GTBB inverter disconnect from the grid, power the back-up load panel and use the power from the HVBF system to power the critical loads in the back-up load panel. It also provides power to the GTBB inverter to charge its battery bank.  If this sounds a bit complicated, well, it is.


Courtesy of Schneider Electric

AC coupling provides the following advantages over traditional DC-coupled GTBB system designs:

  • Retrofit-able with existing HVBF systems (within manufacturer requirements and limitations)
  • Allows for employing the efficiencies of HVBF equipment while achieving back-up power for utility outages
  • Can reduce the number of components used in DC coupling
  • Can reduce losses do to low-voltage aspects of DC-coupled systems
  • Can provide for more flexible and efficient wiring configurations
  • For designs requiring long distances between the renewable energy resources and the balance of system components

As with any innovation, AC coupling has some notable challenges, especially when the design utilizes multiple manufacturers.

For several years, system integrators have completed AC-coupled designs using one manufacturer’s equipment or by using multiple brands of inverters.

SMA pioneered the AC coupled concept with its “Sunny Island” Inverter. Initially built to provide for the creation of microgrids on islands and other non-utility environments. The design lends itself to grid-intertied AC-coupled systems as well.

As shown in the diagram below, SMA’s design allows for multiple HVBF inverter outputs to be combined with the Sunny Island inverter to connect to the utility and have battery back-up.


Courtesy of SMA America

SMA’s design provides for an elegant method of regulating the battery state of charge as long as all the inverters can be networked with cat-5 cable. In this design the HVBF inverters can have their outputs incrementally reduced as the battery reaches a full state of charge. If the distance between the HVBF components and the Sunny Island is too great to network with cat-5 cable, the Sunny Island controls the output by knocking out the output of the HVBF inverters with a shift in the frequency of the inverter’s AC waveform.  The HVBF inverter senses an out-of-spec frequency and disconnects until the frequency is back in spec for five minutes.

This frequency shift method of regulating battery state of charge is often used when different manufacturers’ inverters are used to create the AC-coupled design. This has several drawbacks that we will discuss.

Several other battery-based inverter manufacturers have developed designs for using their inverters with other HVBF inverters to create AC-coupled designs. These include OutBack Power, Magnum Energy, and Schneider Electric. Both SMA and Schneider provide for single manufacturer AC-coupled systems because they manufacture both HVBF inverters and GTBB inverters. This presents the basic advantage of having one manufacturer provide and support the entire AC-coupled design.

OutBack Power and Magnum Energy manufacture only battery-based inverters and therefore require the mixing of manufacturers in AC coupling in order to bring in HVBF inverters.

Both companies provide design information and support for AC-coupled designs.

Schneider’s regulation

With Schneider Electric’s AC coupling, the battery is regulated by the frequency shift method. Schneider itself recognizes the drawback of this method in its AC-Coupling Application Note (see appendix): “Unlike its normal three-stage behavior when charging from utility grid, the Context XW does not tightly regulate charging in a three-stage process when power is back fed through AC inverter output connection to the battery. In this mode charging is a single-stage process, and the absorption charge and float stage are not supported. Charging is terminated when the battery voltage reaches the bulk voltage settings, which prevents overcharging of the batteries. Repeated charging of lead acid batteries in this way is not ideal and could shorten their useful lifetime.”

This can be improved by employing a diversion load controller added to the design.  The diversion load controller will limit the battery voltage by “dumping” excess power into a DC load during times of excess generation for the PV system. While this re-introduces the 3-stage charge regulation into the design it negates some of the benefit of AC coupling because it re-introduces the cost of a charge controller and adds the cost of the DC diversion load(s).

Magnum’s regulation

Magnum Energy also provides for frequency shift method battery regulation but in their White Paper titled “Using Magnum Energy’s Inverters In AC Coupling Applications” (see appendix) they indicate that frequency shift regulation should only be used as a back up to the employment of a diversion load controller. They are developing an innovative addition to their product line the ACLD-40, which will provide for diversion control using AC loads. One aspect of using diversion load controllers is that DC loads are often difficult to find and expensive. Magnum intends the ACLD-40 to be a solution to this issue by allowing the use of more common AC loads for diversion controlling such as AC water heaters or air heaters. This product is under beta testing at this time and is due for release in late 2014.


OutBack’s regulation

OutBack Power’s design provides for frequency shift method battery regulation. The disadvantages to this method can again be overcome by the introduction of a diversion load controller and this comes with the same issues as with the other manufacturers.   OutBack Power’s AC coupling white paper discusses both on and off grid applications for AC coupling (see appendix).

Disadvantages to AC coupling:

  • Frequency shift methods of regulating the battery state of charge are coarse and may create significant power loss if there is a miss-match of equipment leading to nuisance tripping of the HVBF inverter
  • Battery optimization may not be possible without re-introducing a charge controller as a diversion load controller
  • Complexity in systems mixing manufacturers can create systems that are difficult to operate
  • Care must be take not to void warrantees by using equipment that is not designed for this application


In many ways, AC coupling is a good tool for working with both the difficulties of retrofitting battery storage in existing HVBF systems and systems with long distances between resources and loads. As with any innovation in this field, be sure to get the right design and make sure that the application does not void product warrantees.





Posted in Geothermal and Solar Design and Installation Tips, Solar Photovoltaics | Tagged , , , , , | Leave a comment

Three Keys to Developing Bankable Solar Projects – Lessons from Developing 150 MW+ of Solar Projects

Thanks to Chris Lord and Keith Cronin for providing all of the insights in this article. Chris and Keith teach our Solar Executive MBA course (the next session starts on September 15th). Together, they have advised investors, owners, and other developers on more than 150 MW worth of distributed generation solar projects. While there is no standard for defining bankable projects, I trust their on-the-ground experience to provide these insights.

Introduction – Why is Bankability Important?

The majority of large-scale US solar projects are done through power purchase agreements. The key to power purchase agreements is having investors who buy into these projects. As SREC prices and policy continue to fluctuate while project IRRs and installed costs continue to drop, project investors are most interested in investing in bankable projects that have good returns and minimal risk.

One of the most common question we get in our Solar Executive MBA course is, “How can I build a bankable project?”

In other words, “How do I structure a project so it’s very easy for an investor to want to invest in or purchase the project?” For commercial solar, the answer is of course “it depends” because there are so many moving variables that go into projects and everything is negotiable.

In this article, we’ll define and go into three keys to developing bankable projects. Then we’ll go into what developers need to keep in mind to stop wasting time on chasing bad projects.

This article will be useful to a professional who needs to get good at or keep up to date with best practices for financing mid-market solar projects. The goal is to go deeper than most articles on the Internet, but it will be impossible to provide the deep dive necessary to make you an expert. If you have any questions about the content, please leave them in the comment section of the article.

More Reading

If you’d like to get more resources on the subject, here they are. We will reference all of these articles in the article, but I wanted to provide a simple list for ease of use.

Article Outline and Learning Objectives

The article will be split into four major sections. After reading this article, you should understand the most important factors that go into developing a bankable solar project. The goal is that you’ll become familiar with these variables and will be able to start screening your existing projects to find the bankable ones and will also be better at screening new potential customers.

Defining Bankability

Technically speaking, bankable projects mean investment-grade projects. These are the projects that are the economically strongest and most reliable (i.e. low-risk) projects. These projects are able to win the most conservative and lowest-cost capital.

Economically strong is defined as a project that uses reasonable or conservative assumptions and documented facts to create a healthy economic cash and tax flow that comfortably hits or exceeds the investors’ target IRR.

Economically reliable is based in part on the strength of the developer/construction entities and also on the confidence of any state subsidy. But is also heavily based on the quality of the design and construction of the project. Investors want to know that the project was built to a rigorous standard so its economic performance may be reliably predicted to actually generate the forecast numbers in the pro forma over a twenty or twenty five year term. That is a long time by anyone’s standards. Imagine if a project were a car. In twenty years, do you expect to be driving the same car you are driving today? Probably not. A high-quality project is not a project built to meet minimum performance standards at the lowest possible cost.

  • Economic performance is most commonly addressed by establishing production guarantees. Oftentimes, investors will negotiate for the developer or EPC to guarantee a certain level of project, this is especially true if the equipment is being finance under a PPA and not a lease.
  •  The other item that impacts economic performance in modeling is the use of P50 vs P90 production levels. Investors will typically want to use P90 production numbers because they are the most conservative. Read more about production modeling 101 here.

First Key to Bankability  Understand How Investors Evaluate Projects

Solely on a 20-year discounted cash and tax basis.

Investors value a project based solely on the cash and tax benefits that will flow from the project. Similar to how you might value an annuity, they are paying good cash for the right to receive the cash and tax benefits from a project. Project the annual benefits over a twenty or twenty five-year term, and discount each year’s value back to the present using your target return rate.

This valuation method creates a problem for developers.

  1. Developers’ first instinct is to cut construction costs to the bone. Why not? After all, the difference between the development/construction cost and the sale price to the investor is all margin for the developer.
  2.  Projects are also judged on their quality – performance and reliability. In fact, successful developers have learned to fight the instinct to indiscriminately attack costs and to focus instead on managing costs intelligently with an eye to longer-term value.
  3. The key learning here is that it’s not the project with the lowest installed costs that wins, it’s the projects with the highest returns. This takes into consideration installed costs, the amount of power that an array will provide, and the confidence that the installed costs and power production will be very close to what’s expected.

A Second Key to Bankability – Have a Strong Economic Model

It’s key for commercial solar projects to have a comprehensive economic model. You need to know what kind of return you are really offering your investors before you show them the project.

It’s okay to start with a simple model for initial project screening and early development, but the sooner you move to a comprehensive and robust model the sooner you know where your project’s strengths and weaknesses are so you can then develop the project accordingly.

It is extremely important to use reasonable assumptions on all variables of the project economics. This includes: installed costs, PPA price, sales tax, property tax, interconnection costs and timelines, and SREC prices.

It’s important to lock in the “knowns” or “facts” of a project. This means variables that are documented and that you are close to 100% certain of their value. Be clear about which variables in the project are known and unknown.

Comprehensive and accurate documentation of variables is essential. The fastest way to lose the trust of an investor is not properly performing your due diligence by gathering information on all the necessary variables or not accounting for them correctly. For example, interconnection costs and real estate are not eligible for ITC and MACRS depreciation. Did you remember to exclude them? Not properly discounting the ITC is the single most common modeling mistake, even in large projects.

A Third Key to Bankability – Weighing Capital Costs Against Operating Expenses to Maximize Project Returns

This links directly to having a proper economic model. In your model, you need to know and understand what saving $1 on the operating side means relative to $1 on the capital side. The impact is different and depends on the facts.

  • For example, on a 5 MW (AC) project on the East Coast, cutting the construction cost by $.10 a watt (or $500,000) raises the project IRR by approximately 0.4%.
  • On the same project, cutting $6,500 of annual expenses by finding a lower-cost property raises the IRR by almost the same amount. The greater impact comes from the recurring impact of the lease rate reduction. In other words, the $6,500 is realized every year over the term, not just the first year.
  • Effective and valuable cost-cutting involves weighing capital cost reductions against operating expenses – and this is where good development and good design can help.
  • Energy efficiency in buildings offers a very easy way to see this trade-off. Imagine a developer looking to build a commercial office space. The developer might consider a highly-insulated and energy-efficient window solution but reject it because the cost is “too high.” Instead the developer goes with a very cheap but not very efficient window solution. After the building is completed, the operating cost of the building with lower efficiency windows is higher because of the additional energy required to warm and cool the interior space. Had the building owner gone the other way, the capital cost would have been higher, but the operating cost would have been lower. The trade-off is never an easy one to make, but in the building example, if the tenant is not the developer/owner, then the trade-off involves shifting costs from capital (owner/developer) to the tenant (who pays the energy bill).
  • A solar project requires the same trade-offs on the design, choice of materials, and construction. And, as we saw in connection with knowing your model, the impact of changes can vary considerably depending on whether you are cutting capital costs or cutting operating costs. In both cases, you need to know how the IRR is impacted and whether cutting the capital cost makes for a lower life cycle cost.

The Developer’s Perspective     

After Chris Lord provided this advice on the modeling and legal aspects of developing solar projects, I asked Keith Cronin a simple question, “This advice seems so clear, why are developers not following it? What are they chasing around bad projects? What advice do you have for them?”

Here’s an excerpt from his response:

Developers around the globe all want to seize opportunities in the solar industry, as they see gold in their eyes. This has been happening for almost a decade now in various iterations. Small and large developers are always looking at incentives, pulling out their spreadsheets, and eagerly looking to secure properties to park these opportunities on them.

What developers often overlook is the identification of a good versus a bad opportunity. As Chris Lord points out, determining bankability is essential. Discovering that a project can’t be financed is a large source of disappointment for developers after they’ve invested hours of time chasing deals.

This stems from a host of variants, but these are the most likely primary offenders:

  1. Developer runs into unforeseen conditions at a project’s location and the additional costs make the investment economically unattractive.
  2. Cost for construction and interconnection delays decreases project returns. Projects in the Hawaii market that have been involved in the FIT program have experienced 18-plus month delays from a host of parties involved in a project. For example, if you look at a 500kW AC PV system producing $23,000 per month in revenue, how many developers can afford to lose $400,000 during that time period, and how many investors have that level of patience? If you look at Chris’s example with capital expenditure versus operational expenditure and how this impacts project returns, any hiccups with construction delays can substantially decrease project returns.
  3. Uncertainty around interconnection makes some projects impossible. If programs become oversubscribed and circuits on the grid become saturated, how do you explain to investors that you will not only see additional cost overruns, but the likelihood of waiting until the infrastructure can be modernized to impact the project timeline?

As Chris Lord points out, the cost of a project, versus the recurring costs for land, insurance, taxes, leases etc., should be carefully scrutinized. As developers, we all want to build a project for less than what we planned for. What is the best strategy for the short term and long term?

  1. It is advisable not to cut corners on solar equipment because arrays have a useful life of 20 to 30 years. Make the long-term investment and build that into your budget. Be prepared to tell the investment community “why” your costs are higher and they are usually thankful for the insight and your long-term thinking.
  2. What is your O&M strategy? These numbers fluctuate radically. It can be anywhere from $10 per kW to $25 per kW per year on larger scale projects. The bigger question is what is included in this service that will be provided and what isn’t? Remember, the PV system’s output will not go up over time and only go down with degradation, so plan for the impacts of time on a system and understand how your investors are looking for less lumpy returns and more stable forecasts.
  3. Bundling other complementary services offers a unique angle on getting your project to the finish line. With the cheap cost of capital, investors are looking for low-risk returns, and adding in energy efficiency will stabilize returns for the investment community. Access to this money today is easy to find. In places with high-energy costs, it makes the amalgamated deal more attractive and often can give a developer the margin they were originally looking for at the onset.

Project risks often burden developers in the early stages of a project’s introduction and inception. Engineering, permitting, site control, and negotiations with landowners all consume a lot of in-house resources. Getting better at selecting projects that have a higher probability of being developed requires experience and knowing the market you’re entering. Finding local partners that can help you traverse the nuances of the market is essential in maintaining your expectations as well as the expectations of the investment community.


This article outlined the knowledge you need to develop and screen existing and new bankable solar projects. You now know what you need to keep in mind to stop wasting time on chasing bad projects.

The most important factors that go into developing a bankable solar project:

  • Understanding how investors evaluate projects (solely on a 20-year discounted cash and tax basis)
  • Strong economic modeling
  • Weighing capital costs against operating expenses to maximize project returns

If you have additional knowledge to share, please leave it in the comment section below.

Learn More

If you’d like to get more resources on the subject, please review the following resources (they were also included in the article). Please leave comments in the comment section below if you have questions or additional knowledge to add!

Posted in Solar Photovoltaics | Tagged , , , , | 1 Comment

Modeling Solar Production Risk 101 – An Introduction to P50 and P90 Production Levels



This article is part of a series of interviews, tutorials, and definitions around commercial solar financing that is leading up to the start of our next Solar MBA that starts on Monday September 15th. In the Solar MBA students will complete financial modeling for a commercial solar project from start to finish with expert guidance. The class is limited to 50 students, but there are 30 discounted seats. 

Financing a large commercial solar project is about understanding, controlling, reducing, and communicating risk and uncertainty.

Because solar has a variable energy source, the amount of power that an array will produce, and thus the value of that power, is highly variable and needs to be understood to finance a project. As projects get larger, more due diligence is required to understand and evaluate the potential solar production of an array.

Solar production estimates are based on a number of factors. Some factors can be controlled and modeled with a high degree of certainty and others are closer to guesses about the future. Because these are guesses, we need to state a confidence level for each guess.

The confidence level of the amount of energy a solar array will produce is measured in P50 and P90 production levels.

This article will be useful to any solar installer who sees commercial solar projects, and specifically the financing of those projects, as critical to the success of their business. For larger projects, PVWatts won’t cut it. You’ll need to understand the amount of potential revenue a solar array can generate, and your confidence in its ability to generate that revenue,  in order to get investors to buy into the project.

In this article we will explain:

  1. The general types of production risk and why there is such a huge focus on solar radiation levels.
  2. The definition of P50 and P90, how it’s graphed, and the impact of weather variability.
  3. The potential impact of revenue expectations.
  4. What is critical to understand about P50 and P90.
  5. Further reading.

Let’s dive in.

1. Types of Risk

There are many ways to describe the risks associated with a solar array. In general, you could put them into two buckets: “construction risk” and “operating risk.”

From an investor’s perspective, construction risk is any source of risk that happens before COD, when the system is not operating. This can include site risk, site control, interconnection risk, EPC and construction risk, and more. An entire article could be written on those topics. For the most part, construction risks are about understanding and controlling the cost and time required to build the array. There are some obvious factors during the engineering and installation of the array that can have large impacts on the potential production of the array.

Operating risks are the risks associated with running the facility and generating revenue from the production of energy. These can still include some site and equipment failure or warranty risks, but, assuming those are controlled for, the major risk after a solar array has been constructed is how much power it will produce.

AWS Truepower published a report about reducing uncertainly in solar energy estimates in which they rank ordered the factors that have the largest potential impact on solar production estimates. That graph is below.

Screen shot 2014-08-06 at 9.39.08 AM


As you can see, “solar resource uncertainty” is the single largest item that can impact total solar power production based on their analysis.

David Park from IEEE published a similar analysis. He rank ordered the impact solar radiation, climate, module model, inverter model, aging, and system derate can have on expected array production versus estimated production and found that solar radiation, climate, and radiation and module model explained the largest amount of production variability.


The value of energy produced by a solar array is a function of two items: how much energy is produced and the value of that energy. The value of that energy can be based on a number of factors: the kWh rate it’s offsetting, any net-metering laws that are in place, the negotiated PPA rate, potentially demand charge reductions, any production-based incentives, and more.

Given that we cannot predict with 100% certainty the amount of solar radiation that will hit an array over any given period of time, to understand and communicate the potential solar resource we use P50 and P90 production levels of an array.

While these production estimates rely to some degree on system design and siting, the main variable is weather.

2. The definition of P50 and P90 and how they are graphed

In P50 and P90, the P stands for probability.

P50 means there is a 50% chance in any given year that production will be at least a specific amount. If an array has a P50 production level of 500 kWh, it means that on any given year there is a 50% chance that production will be AT LEAST 500 kWh.

P90 production means that there is a 90% chance that in any given year production will be at least the specific amount. This means that there is only a 10% chance that production will be lower then the stated amount. If any array has a P90 production level of 400 kWh per year, it means that on any given year year there is a 90% chance that production will be AT LEAST 400 kWh.

Here’s a graph of P50 and P90 production estimates from David Park’s report.


For any statistics geeks who are reading this, it will look very familiar. What he’s doing is graphing the variability of the mean in a confidence interval, in this case one standard deviation is 12.5%. P50 is the mean and P90 is a little less than two standard deviations (remember that two standard deviations is 95%) from the mean.

3. Weather Data Variability and the Relationship Between P50 and P90

Because the variability of solar production, and thus the difference between P50 and P90, is largely based on the variability of weather, extensive weather analysis must be performed to calculate these values.


The above graph comes from the report by David Parker and illustrates how weather recordings over a specific period are used to determine the variability of irradiation for a specific location.

What this means for solar production is that areas that have less sporadic weather changes have closer P50 and P90 values. Statistics geeks, lower variability means a lower standard deviation across the distribution of solar irradiation values.

If you look at the two graphs below, the top graph is an illustration of an array that has highly variable weather characteristics while the bottom graph displays an array with more stable and predictable weather.


4. Potential Impact of P50 and P90 Production Estimates on Revenue Potential

The difference between P50 and P90 production levels in areas with moderately variable weather can have large impacts on the assumed production for an array.

If we want to use the example from our first graph:

P50 production was: 32,413 kWh

P90 production was: 27,228 kWh

That’s a difference of 5,185 kWh. P90 production estimates are 15.9% lower than P50 values. 15.9% is a lot!

Assume the value of a kWh is $.15 per kWh

P50 production is expected to be: $4,861

P90 production is expected to be: $4,084

If we assume that each kWh is worth exactly the same amount, this means that the value of the power produced would be expected to be 15.9% lower if P90 was used compared with P50. However, we can be much more confident that every year we’ll hit the P90 production levels. This is why investors signed into a PPA tend to favor P90 production levels if they are being paid with power production.

5. What’s critical to understand about P50 and P90?

  • P50 and P90 production levels are hard to determine with software models.
  • For larger projects, an engineering firm will work to perform this analysis.
  • P90 is more conservative, so investors will focus on this amount. P50 is less conservative, so developers tend to focus on this.
  • The greater the variability of weather in a specific area, the greater the variability between P50 and P90 because solar radiation levels explain the majority of the variability in production.
  • Investors will be most concerned with production levels in legal structures where their returns are based on the production of the system. In lease structures, the investors will be less concerned with production because the payments are hell or high water payments.

6. Further Reading on Production and Risk Modeling


Posted in Solar Photovoltaics | 1 Comment

Troubleshooting Condensing Boilers in Hydronic Systems – What is the System Doing?

This a guest post from Roy Collver. Roy is a condensing boiler expert. Here’s what John Siegenthaler, author of “Modern Hydronic Heating,” says about Roy’s work: “When I have a detailed question about the inner operation of a modulating / condensing boiler, Roy Collver is the first person I contact. The investment in Roy’s HeatSpring course is a fraction of the cost of a single mod/con boiler, but it will teach you concepts, procedures, and details that will return that investment many times over.”

Learn from Roy

  • Free. Roy is teaching a two-part free course on how to sell mod-con boilers. The second live lecture is happening on Wednesday, July 30th. Sign up for the free mod-con course here.
  • Paid: Roy Collver teaches an advanced 5-week course on mastering condensing boiler design in hydronic systems with the folks at HeatSpring. If you need to increase your skills and confidence around selling, quoting, designing, setting up controls, or troubleshooting condensing boilers in new construction or retrofit applications, this course is for you. Each session is capped at 50 students, but there are 30 discounted seats. Get your discount and sign up for Condensing Boilers in Hydronic Systems.

Enter Roy…

Understanding the Simple Basics

Cold weather is never too far away in most parts of North America. Be ready when it hits, and review the basics of hydronic system operation so you can quickly locate the problems that always come up. When you approach an operational hydronic system it will exhibit one of the following six states. Quickly understanding what you are dealing with will greatly reduce head-scratching time and point you in the right direction. Standing slack-jawed in front of a boiler with no clear path to determining what is wrong is very uncomfortable and a waste of time. Confidence is a key factor in successful troubleshooting, and to be able to indicate to a customer what the BASIC problem is right away buys you time to be able to work the problem, find out the SPECIFIC cause, and fix it. Using this guide as a quick reference should help speed the troubleshooting process along.

Hydronic systems are all about Delta T (the difference in temperature between the heating fluid, the system components and the surrounding air and objects). Heat always travels to cold, and if heat is not added to the heat transfer fluid (usually water), the fluid and all of the components in the system will eventually cool down to the temperature of the surroundings.



The boiler is on and the hot combustion gases create a large Delta T between the combustion chamber and the water in the surrounding heat exchanger. Because heat travels to cold, the water heats up. The circulation pump moves the hot water through the distribution piping to the terminal units. The terminal units heat up and a Delta T develops between the hot terminal units and the colder air. The air will get warmer at the expense of the water, which cools slightly. The cooler water circulates back through the system back to the boiler where it is heated up again. If the heat going into the boiler is more than the system can use, the water will continue to get hotter until the boiler cycles off on its operating control. The temperature difference between the water leaving the boiler and the water returning to the boiler will be “normal” for the system (usually 15°F to 40°F depending on the load and system design).

noflowThe boiler is on, adding heat to the water, but for some reason the hot water is not circulating through the distribution piping to the terminal units. The terminal units will cool down to the temperature of their surroundings and a “no heat” condition will result. The water in the boiler will continue to get hotter until the boiler cycles off on its operating control or internal high limit control. The supply and return piping near the boiler will be close to the same temperature.



The boiler is on, adding heat to the water, but the hot water is not circulating fast enough through the system. The first terminal unit may become warm, but because the water is moving so slowly, all of the usable heat is transferred out of it before it gets very far. The last terminal units do not become warm enough to heat the space and a “not enough heat” condition will result. The water in the boiler will continue to get hotter until the boiler cycles off on its operating control or internal high limit control. There will be a large Delta T between the water leaving the boiler and the water returning to the boiler. (The supply will be a bit hotter than normal, but the return will be much colder than normal.)


Continue reading

Posted in Building Efficiency | Tagged , , , , | Leave a comment

New Massachusetts Solar Bill H.4185 Would Destroy Community Solar Potential In the Commonwealth

This is a guest post from Sam Rust from SRECTrade about new solar legislation in Massachusetts.

(Editor Note: A note about the importance of community solar for lowering customer acquisition costs, something EVERYONE in the solar industry cares about. Everyone is talking about lowering customer acquisitions costs and soft costs and community solar has the potential to instantly drop acquisition costs by 50% to 80% for solar companies offering roof mounted and community solar projects. Why? It’s simple math. If you had 100 solar leads, a good conversion rate of leads to customers would be 10%. This equals 10 customers. Here’s the thing, in order to find 10 customers that WANT to invest in solar and HAVE a good roof, you must bump into 3 to 4 people that WANT solar but DON’T HAVE the roof space. If those 3 to 4 people could become community solar customers, then the conversion rate of those 100 leads would become 30% to 40% instead of 10%. This would then drop the acquisition costs because you’re getting more customers with the same marketing spend. Food for thought.)

Enter Sam Rust.

In 2013 Massachusetts was ranked 4th, behind California, Arizona, and New Jersey for most solar installed. Despite this success, legislation, officially known as H.4185 (An act relative to net metering), is pending at the Massachusetts State House that could drastically change the direction of the Massachusetts solar industry. Touted in the media as successful compromise between regulated utilities and the solar industry, H.4185 might be more of a step back, than a step forward. The bill could pass in both the Massachusetts House and Senate before the end of the legislative session on July 31st, despite the opposition of many solar owners, installers, and representatives of the community solar movement.

Here’s a short explanation of how we got here and what H.4185 is.

Currently there are limits on how much Massachusetts solar capacity can qualify for net metering in each utility territory. These limits, which only apply to larger solar facilities, are nearly maxed out for each utility and prevent the Commonwealth from meeting Governor Deval Patrick’s 1,600 MW by 2020 solar goal.  H.4185 would remove the net metering limits and put in statute Governor Patrick’s 1,600 MW target in exchange for a radical adjustment in the structure of Massachusetts solar policy of which the primary adjustments are:

  • The removal of annual capacity restrictions on large “solar farm” projects
  • The creation of yet-to-be defined minimum electric bills for all ratepayers
  • The reduction of the virtual net metering rate from compensation at the retail rate to the wholesale rate of electricity
  • A limit on the size of behind-the-meter projects to 100% of the on-site load
  • A transition away from the successful market-based SREC program to an unknown program managed by the Department of Public Utilities
  • Transfer of all of the “environmental” attributes of solar arrays to the utilities

In translation, H.4185, a bill that is ostensibly about net metering would remove or weaken most of the policies that have made the Massachusetts solar industry so successful. It is a bill that exchanges a set of known, highly successful policies, for a new set of untested policies.  The bill has not yet passed and many stakeholders are calling amendment language that would remove most of the major policy language in exchange for an incremental increase in the net metering caps and a formal commission to be convened next year to review the more contentious aspects of the legislation. This more cautious approach would stabilize an already jittery Massachusetts solar industry and ensure that all stakeholders are at the table the next time net metering limits need to be addressed.

How this could negatively impact solar installers.

  1. Anybody working to do community solar will be negatively impacted because the VNM credit is being reduced
  2. H.4185 removes the protections in place under the SREC-II program for incentivizing distributed/ rooftop/ carports/ general behind-the-meter projects
  3. The declining block incentive program will be set at the DPU, rather than at the DOER. This means that installers will need to lawyer up and deal with the regulated utility lawyers in order to argue for favorable incentive targets. Solar in Massachusetts goes from a decentralized system, where everyone and anyone can participate in the rule making process to a system where the big player have the negotiating advantage
  4. The utilities receive all of the attributes of the solar, including the RECs and will be able to lead the discussion on monitoring and other equipment requirements. This reduces the possibility for innovation in the solar space regarding capacity markets, battery storage, voltage regulation etc
  5. The minimum bill imposition will hurt anyone with a low electric bill, which means smaller projects will be most affected by the minimum bill
  6. Anybody doing business in Muni territory is left out. Currently the SREC program covers Munis
  7. Above all else this just adds more complication to the system. We just spent a year implementing SREC-II and now we have to work on implementing another program for which installers will need to fight to be part of the process for negotiating the declining block targets and minimum bill. This just adds more uncertainty, which is bad when you are trying to mature an industry.

WANT TO HELP? Contact Sam

First. Here is Sam’s email address:

Send him an email and he’ll figure out how you can help.

Massachusetts voters are encouraged to research this bill further and to contact their state legislators. Here is a link to a site that makes legislators searchable by zip code.

For more information please read this well written opinion piece in Commonwealth magazine and visit the Facebook page for the Massachusetts Stakeholders for Competitive Solar or


Posted in Solar Photovoltaics | Leave a comment