Solar is truly in its teenaged years, and that means that, my friends, it’s time to talk about midlife and of life strategies like repowering solar projects.
Teenaged years?! But solar’s been around for a hundred years you say……
True, but as an industry that provides true impact to the energy grid solar is just starting to come into its own.
And don’t get me wrong I’m not looking back at the last decade or so and saying what we had done up to this point was bad or wrong (although if you are interested in taking a stroll down memory lane – do I ever have an article for you). We’ve made so much progress, and grown fast, but it is time to start planning ahead.
To run my own metaphor; we’ve seen incredible growth spurts in solar over the last few years, and I think we’ve faced an identity crisis. We went from being a remote application, to becoming part of the global grid. Now solar faces the need for some self awareness and introspection, it needs to look at all the “in the moment” design and build decisions with a critical eye, and start thinking about being part of the future energy mix.
Repowering Solar Plants: What to Consider
The driving force for that learning and growth in my mind will come from our repowering initiatives. When I talk about repowering solar projects, I’m talking about replacing or adding equipment to improve performance and/or extend the life of the facility. I know, we can debate what qualifies as repowering vs maintenance, but that is a whole other article.
Happily, I have had a lot of discussions about repowering lately. I have moderated 3 panels on repowering over the last year, and seen it crop up in countless other discussions and panels. It seems repowering is definitely on the minds of the industry.
It’s not surprising either, we are starting to think about the facility and how to extend its life or improving its performance.
A 25 year life span was hard to imagine when we first built these projects, but now we’re at the midway point. We’re starting to see the end of the facilities’ life cycle, and to see the impacts of 10 years worth of wear and tear have on our equipment. As one of the panelists on a session I recently moderated in San Francisco aptly put it, “It’s no longer just a line item on a financial model.” This is real money we need to spend, and a real problem we now need to solve. Whatever you take away from reading this article, I hope you remember these few things about repowering solar projects:
- It’s more than just the equipment
- Time has a cost
- In retrospect we should have considered this in design
It’s more that Just the equipment
Most large facilities built over the last 10 years have a single line item for any form of equipment replacement in its lifespan. This line item is usually a thumb in the air, so to speak – a rough estimate of what we thought it would cost to replace half of the inverters on site in 10 years. And that field typically predicts spending that budget in year 10 of the financial model.
We now know that the scenario I have just described is not the case in real life.
In 10 years, technology has moved on, and models have been redesigned. The unit might be cheaper, but is it still compatible if it’s 1500 volts in a facility only designed for 600? If the connectors have changed, or if we have to rebalance strings, or the port is now on the other side of the machine. Any and all of these changes means a redesign is probably necessary. After that’s done, you’ll need to hire a crew to do the work, and you may have some permits and certifications that need to be managed. You may even need to sign amendments to your contracts. All of this is well above and beyond the future cost of an inverter anticipated 10 years ago. Of course, this gets more complicated for more complex pieces of equipment like solar panels or combiner boxes.
Time has a cost
I’m honestly a bit surprised we missed this one, in retrospect. The system will be down. It’s such an obvious statement, but your original projections probably didn’t take it into account. It might be some time before it’s back up and running, and it should be considered a cost, or at least as a decreased production for a material amount of time. Think about how much of the system you are required to isolate, and how long it takes to refurbish an inverter. Especially if there are any other restringing or rebalancing efforts that need to occur in tandem. Then commissioning and certification. 3 days, or 5 days if we’re being optimistic, per inverter. It adds up.
What model can withstand a 5-10% product hit for 2 months without someone worrying? This is definitely not part of the plan.
In retrospect we should have consider this possibility in the design
When you build these solar projects, you start with a blank slate by definition. Your open space is likely pretty flexible, and you can plan your construction appropriately to deal with equipment and their space needs and the limitations of the lot. Unfortunately, after it’s built, access isn’t always as easy as it once was. During development, you had ample space between the piles to drive a loader, but not now, because the rack and panels overhang. That trunk line that passes under the rows, now has a fault, and you need to figure out how to dig it up. That inverter transformer set you built off site and shipped as a complete package has to be removed and shipped back so that it can be dismantled and fixed.
We need to remember to design and build for the future. We need to consider the ability to remove and repair equipment, and to facilitate the ability to maintain and manage the equipment. Putting a plan in place that reflects the reality of repowering solar projects is essential.
So where to now?
To start, plan 20+ year lifespans for your current and future solar projects. Once we hit our stride, we’ll see predictability, set standards, and maybe, in the future, have this same conversation once more, focused on 40 year life spans. We can refinance with confidence, and expand, and improve facilities to get the most out of our systems without tripping over decisions made in the past.