Episode 43 Transcript: Enphase and a Pool of Energy at Every Home

Molly Wood Voice-Over:
Welcome to Everybody in the Pool, the podcast for the climate economy. We dive deep into the climate crisis and come up with solutions. I'm Molly Wood.

This week I’m excited to talk to one of the success stories from what is sometimes called Cleantech 1.0 in the climate tech and investing world—a period between 2006 and 2011 when about $25 billion worth of venture capital was invested in clean energy startups. It was a bubble followed by what some people considered a bust, but several of those companies are still around and have made back those billions many, many times over—think Tesla, Sunrun, who was on the show, Beyond Meat, and my guest today, Enphase Energy.

A now public company that created all-new microinverter technology for controlling and optimizing rooftop solar panels, Enphase is now innovating in battery technology and smart home energy systems to put affordable, renewable, and resilient power everywhere possible.

In my favorite way … de … CENTRALIZED.

Raghu Belur:
My name is Raghu Belur. I'm the Chief Product Officer and co-founder of Enphase. Enphase is a global energy technology company. We are a renewable energy technology company with a focus in solar and batteries and EV chargers, etc., in order to deliver a very sophisticated home energy management system.

Molly Wood:
Give me a little bit of the background, if you wouldn't mind, of the company. How was it developed, and what was the kind of central thesis?

Raghu Belur:
Yeah, that's a great question. You know, as all entrepreneurs, particularly here in Silicon Valley, you're always looking at a problem and thinking, "Hey, I can do this better, right? I can do it differently, or there's a better way to do it."

Martin and I, who is the other co-founder of the company, the two of us had come out of different startups. I began to notice how technology evolves, where in the long run, everything moves to a decentralized or a distributed architecture. In network parlance (because networking is where we came out of), this means the endpoints of the network become more intelligent, and they start making autonomous decisions.

Molly Wood:
You…

Raghu Belur:
Contrast that to what a centralized topology would be, where the endpoints would be pretty dumb, and all knowledge is centralized and decision-making is done in a centralized manner. That's not how the technology world evolves. If you look at mainframe computers to PCs, data centers, or smartphones, these endpoints are becoming extremely intelligent.

There's a common thread that runs through the examples I've given you, and that thread is it's all about digital and software. That's the foundation that we came off of, that the world moves in that direction. For me, this was my fifth startup. Martin and I had started one before, and Enphase was his sixth startup. Our whole career had been startups; it's kind of how we were trained.

When we started looking at energy, particularly solar, we realized it was ready for disruption because it was being done in a centralized manner. Solar itself, both residential, commercial, and utility-scale, was done in a centralized manner, meaning power conversion was being done through a big-box inverter.

Molly Wood:
Yep.

Raghu Belur:
You aggregate all the solar panels, feed them into one box, and that box takes the output of the solar panel and converts it into AC that's usable in your home. Our view was that if our thesis holds true that in the long term everything moves to a decentralized or distributed architecture, and it does so because of cost, performance, and reliability reasons, this needs to be disrupted. We need to move from centralized power conversion to fully distributed power conversion, distributed energy generation, which is through microinverters.

Of course, we had to reinvent the microinverter quite candidly using digital technology by building our own custom ASIC and leveraging a lot of software. We realized that would be the recipe for success if we could effectively do that. But we had to come up with some really major…

Molly Wood:
Mm-hmm.

Raghu Belur:
…technological breakthrough to achieve that because what you're trying to do is build something that's going to go up on the roof, in a very harsh environment, and it has to last for 25 years. How do I make a piece of power electronics that can last on the roof for 25 years?

Molly Wood:
Mm-hmm.

Molly Wood:
So I'm going to ask you for some definitions for people who aren't too familiar with this. But first, just at a foundational level, I want to sort of put a fine point on the fact that you set out not to disrupt just the energy industry but disrupt the renewable energy industry as it had already sprung up.

Raghu Belur:
Yeah, because our view was if you take a step back and look at the way the energy business is evolving, it has to become renewable. It's exactly right. We had already done a pretty horrible job of screwing up our planet, so we knew that it had to evolve.

I remember it used to be called "alternate energy." It's not alternate energy anymore. It's renewable as mainstream energy. So we started looking at it, saying, "Within the energy infrastructure, what is it that we can influence?"

We're not going to go into the centralized generation business or reinvent fossil fuel generation technologies. We were focused on what was cutting-edge.

And even there, there was a nuance. A lot of people at the time were looking at improving the solar panel technology. Nobody was paying attention to the inverter, assuming, "Well, inverters have been around for 40 years; what's there to disrupt?" And that was perfect for us because, no, if you think about it differently, if you think outside the box, it's actually ripe for disruption.

Not only did we come up with this new way of doing power inversion using a microinverter, but we also built a full system out of it.

Molly Wood:
Right, wait, before you tell me about the full system, you have to go all the way back and tell people what the inverter does and what it was not doing well so that you had to invent the microinverter. We're going to basics here for folks.

Raghu Belur:
Got it, got it. Well, an inverter is a device that takes the output of your solar panel and converts it into energy that's usable by your home, your appliances, microwave, lights, and TVs, because the output that the solar panel produces is not usable. It's called DC or direct current.

We needed alternating current, which is AC. So the inverter takes DC and converts it into AC to make it usable in your home. That conversion process was being done by a big-box device, a centralized device, a big-box inverter. We said, "No, we need to disrupt that. We're going to do it decentralized or distributed. We're going to take this big-box inverter and turn it into little microinverters and put each one behind every single panel."

Molly Wood:
All right.

Raghu Belur:
So what that does is then every single panel and microinverter pair act like independent energy producers.

Molly Wood:
And so it's more efficient. It harvests more of the actual solar energy.

Raghu Belur:
Exactly, exactly. If you look at the challenges of a centralized architecture as it pertains to solar, the output of one solar panel dictates the output of the others because they're all connected in series. It's like a Christmas light problem.

Molly Wood:
Right, oh, the one bulb, no.

Raghu Belur:
One bulb, exactly. And then you're searching around for it because there's no information to figure out where the problem is. That's the fundamental problem. There's your centralized topology problem.

Because of the series connection, the output of one dictates the output of the others. That's the first problem. The second problem is that big-box inverter is processing a lot of power. It's prone to failures, and when it does fail, it takes your entire system down because…

Molly Wood:
Yeah.

Raghu Belur:
…there's only that one thing that's doing all the processing. There's only one inverter; there's no redundancy.

The third problem is that by aggregating all the panels in series, you have to manage what's called very high voltage DC. High voltage DC is esoteric, so it's difficult to install and manage.

And the last problem is high voltage DC can create arc faults and potentially start fires. So there's a safety element as well.

When we came up with this notion of a decentralized or distributed architecture by taking a small microinverter and sticking it on the back of every single panel, we effectively connected the entire plant in parallel.

Molly Wood:
Right.

Raghu Belur:
Now, each panel and inverter act like an independent energy producer, and the output of one is not dictated by the others. So your Christmas light problem disappears. I can now ensure that every single microinverter optimizes the output of its associated panel.

Your entire plant is in parallel, so you yield more energy. The second thing is that because the unit is processing so little power (a big-box inverter processes 5,000 or 7,000 watts of power, while a microinverter processes about 300 watts), you have many of them.

That means you can make the device. It's not a power device; it's a power electronics device. If you build it using digital topology, your own custom chip, and rely on software, you can make that device.

Raghu Belur:
One, the device doesn't fail. And second, there's no single point of failure, meaning if one microinverter fails in a typical home with 20 solar panels and 20 microinverters, 95% of your system is still up and going. So we address the yield problem and the reliability problem.

The third thing is the output of that microinverter is AC, which is used directly in your home. So it becomes a very plug-and-play device. It was simple to design, install, and maintain because we pushed the intelligence deeper into the network. My original comment about distributed architecture—the panel is now extremely intelligent because of the inverter's intelligence. Now I can track precisely what's going on with every single panel.

Molly Wood:
Mm.

Raghu Belur:
The last element is fire safety. I'm not accumulating high voltage. I'm converting the low voltage that each panel puts out straight into AC. So it's arguably much safer. That was the genesis of the thinking.

We did add some more stuff to it. We made it into an IoT device. This was before the term "IoT" was even popular. It was just natural for us to do that. We embedded full bidirectional communications technology into every microinverter.

Not only was there a lot of software running in each device, but it was also communicating bidirectionally to a gateway connected to a cloud.

Molly Wood:
Yeah.

Raghu Belur:
So we had a full communications link between the cloud and every device. We knew in real-time how every solar panel was doing and how every device in your system was performing.

Leveraging the power of IoT, if you wanted to fix something, we could push a software upgrade, just like you do on your phone. We can still do that today, just put software upgrades on this device.

The link is fully bidirectional. Each device tells us how it's doing, and we can tell the device to change its behavior if needed. That was powerful.

This comes back to thinking about and not losing sight of the big picture. The big picture here was not just about a home; it was about the entire energy infrastructure.

Molly Wood:
Yep.

Raghu Belur:
I joke about this and say if Thomas Edison, Nikola Tesla, or Westinghouse were walking down the streets of Oakland, California or Fremont, California and looked up at the energy infrastructure—the poles, wires, substations, and transformers—they would recognize all elements of it.

This infrastructure hasn't changed in 125 years. It used to be the marvel of the 20th century, but things aren't the same anymore.

Our big-picture vision, which we never lost sight of, was: How do you disrupt the entire energy infrastructure?

Molly Wood:
Yeah.

Raghu Belur:
And that notion of decentralizing a system within the home applies to the entire energy infrastructure. How do I decentralize the entire energy infrastructure?

Instead of having a big central gas-powered plant, oil-fired plant, or coal-powered plant, transmission lines, substation, distribution, and all the infrastructure to come to your house, can I rethink that? Turn it on its head and say the unit of intelligence is the home. Push intelligence into the home, where the home can generate, store, and intelligently consume energy and help the grid out if needed.

This was the notion of reinventing the grid from the inside out. That was our big-picture vision, and it's still true today.

Here's an interesting observation: we never named the company Enphase Solar. We are always Enphase Energy because our view was that we wanted to disrupt the energy market.

Molly Wood:
Yeah.

Molly Wood:
Right. So if it becomes geothermal or backpack nuclear, it could be anything.

Raghu Belur:
It could be anything. Today, we know of a few things. The reason battery evolution comes as a result of changes in the marketplace is because people are starting to electrify.

They're starting to recognize that, "Hey, I want to have a very, very small, if any, carbon footprint."

They're saying, "I want to get off of gas as well. I want to electrify my whole life."

Electrification within the home means electrifying your heating and cooling by going to heat pumps, electrifying your transportation by going to electric vehicles, electrifying your hot water heaters, etc.

rvice the needs of the home when my home is trying to electrify?

Your demand per home will more than double when you electrify. Here's a simple math: an average consumption in the US per home is about 30 kWh per day. An EV will consume about 15 kWh per day. That's a typical 40-mile commute. Two EVs will consume an additional 30 kWh per day.

Add a heat pump for just heating your home, and that's another 15 kWh a day. So now I've more than doubled my consumption. How do I meet this doubling of demand?

Molly Wood:
Yep.

Raghu Belur:
I read somewhere that by 2035, the average consumption in California will go up by more than 40%. That's a massive increase in electric consumption.

There is no way the existing infrastructure can deliver that much energy to the home.

Molly Wood:
Yeah. I feel like it's important for us to be specific about that part of it, that this is about resiliency and reliability and safety, but also about our ability to actually accomplish the energy transition.

Raghu Belur:
Exactly right. With resiliency and all of those are actually table stakes. You must be resilient.

Even that can be made better. I have to deliver now more than 2x the amount of energy I used to consume before. The infrastructure has to support the delivery of that much energy.

I would argue that the existing infrastructure can't do it on its own. You have to be your own generation resource.

You need to be able to generate your own energy, store it, and use it very intelligently.

Because now you're this unit of intelligence capable of producing your own energy, storing it, and using it intelligently, you can also help others who may need the energy when you have a surplus.

Unpowered energy—the future is here today.

Molly Wood Voice-Over:
Time for a quick break. When we come back, more on the future of decentralized energy … and we’re going to get a little ranty on some policy mistakes … from what HAS been the country’s biggest market for rooftop solar …

Molly Wood Voice-Over:
Welcome back to Everybody in the Pool. Let's rejoin my conversation with Raghu Belur of Enphase Energy about decentralizing the grid and how to put the right policies in place to ensure we’re adopting renewable energy at the rate we need to meet our emissions reductions goals—and not screwing it up by taking incentives off the table too soon … CALIFORNIA. Here’s Raghu again…

Molly Wood:
Right. And then talk to me about policy, because you operate nationally and globally. What does the policy landscape look like, and simply put, what would you like to see?

Raghu Belur:
Yeah, you know, policy drives markets. We are in a highly regulated market. Energy is a highly regulated market, and policy makes and breaks markets.

Look at the California Solar Initiative that started in 2007. It's kind of the genesis of the phenomenal growth that California has had in renewables. It is a leadership position in renewables, and it has been a phenomenal, phenomenal example of success.

Contrast that with what happened with the new policy change regarding the net billing tariff, or what's called NEM 3.0.

Molly Wood:
Which we should explain a little bit. It effectively just slashed the amount. Explain it to us.

Raghu Belur:
Sure. Historically, California and the US, in general, has been a net-meter market or net energy mirroring.

What net energy mirroring means is that in the middle of the day, when I have excess solar generation (I'm not necessarily at home), my solar production peaks in the middle of the day. I can export it into the grid, and in the evening, when I come back home or at night, when there is no solar, I can buy back that energy I banked into the grid in the middle of the day. There was a one-to-one relationship. What I sold into the grid in the middle of the day, I bought back from the grid.

Molly Wood:
Yeah.

Raghu Belur:
The new policy says that if you export in the middle of the day, you will get one-tenth of what you were getting paid before or even less. The new policy says that if you export any electron into the grid in the middle of the day, you really won't get compensated at all. You're giving those electrons away for free.

What that means is you really need a battery. So that in the middle of the day, when I have all this excess solar production, I can store all that energy into my battery and then use it when my solar isn't producing anything late in the evening or in the middle of the night.

That sounds great, just add a battery. But the reality is battery technology is still in its infancy.

Molly Wood:
Mm-hmm.

Raghu Belur:
It's evolving in terms of performance, cost, and safety. There's a long way to go.

They got it wrong, quite frankly. The change that the Public Utilities Commission made was too abrupt, way too abrupt. We said there should be a nice glide path into it. We recognize that it needs to happen in the long run.

Molly Wood:
Yeah. Yeah.

Raghu Belur:
It absolutely must happen in the long run; we agree. But don't do it so abruptly. Look at what they've done. I think our California Solar and Storage Association put out a number like 17,000 jobs lost in the renewable business.

Molly Wood:
Mm-hmm, and installation rates.

Raghu Belur:
Plummeted.

Molly Wood:
Well, and worst of all, solar installation rates have plummeted. And so if your goal is energy transition, yeah, I mean, don't get me started on this one. I could give you like an hour of rage on this one.

Raghu Belur:
Yeah, they got it wrong. They got it wrong. I'm in support of it in the long run. It makes sense, but there should have been a transition, a glide path.

Look at what's happening in other countries. In the Netherlands, for example, solar penetration is at 27%. There was a lot of discussion about transitioning away from net metering, and there was a vote two weeks ago. There was going to be a glide path—a well-thought-out five-year plan—but the government said, "No, I don't think you guys are ready for that."

The government said, "We're going to keep net metering for now and revisit it in another couple of years."

What we have to demonstrate in the next two years is that as you step down net metering, the transition will be smooth. Because the fact is we need to transition this energy infrastructure, right? We don't have a choice. We have a 2050 mandate of one and a half degrees C, and we need to make that transition.

These kinds of policy decisions have been really detrimental to that change. There's a lot to be learned from looking at markets around the world.

Molly Wood:
Right.

Raghu Belur:
It's interesting how different markets react differently and act differently. But that's part of the course as it pertains to policy, right?

Molly Wood:
Right. Just out of curiosity, why do you say that long-term change has to happen, that we have to transition off of net metering? Why is that?

Raghu Belur:
Again, I think we need to be a market participant. The argument that a lot of the utilities make is, "Hey, you're using my grid as a free battery."

When you do net metering (in the middle of the day, you export energy into it), there's tremendous locational value for having localized generation. As my demand in the home goes up, it's my solar that's feeding my demand.

The infrastructure simply cannot support the 2.5x increase in demand that's coming. So you need localized generation.

You have tremendous locational value. You don't have the transmission and distribution losses that you have to deal with. If you upgrade your entire infrastructure, it will cost tens, if not hundreds, of billions of dollars.

Molly Wood:
Mm-hmm.

Raghu Belur:
In a volumetric pricing world, that means everybody's rates go up. That's not helpful either.

The balance is localized generation, localized storage, localized usage, with contribution to the network.

I think that is the evolution and the argument.

Molly Wood:
I feel like California is such a beacon in so many ways. We have been a beacon on how to do it right. I agree with you that currently, we're an example of how to do it wrong.

Raghu Belur:
Yeah, they got it wrong. I hope they do something to fix it because it can be fixed. I hope they do something to fix it because we are all in this together, right? We are all in this together.

The utilities have a major role to play. Distributed generation has a major role to play, and we just need to learn to coexist.

Molly Wood:
Yeah. Finally, in our last few minutes, let's talk a little bit about the role of the battery. Obviously, it's becoming more and more clear. But it also seems like the role of the battery is even broader than we initially thought. Talk about the role of the battery going forward.

Raghu Belur:
Yeah, exactly. Historically, people thought the battery was all about resilience.

Even if the grid fails (and the grid has been failing a lot), people say, "Oh, I can switch to my battery, isolate my home from the grid, and build a microgrid to power my home from the battery."

But the role of the battery has evolved. One of the challenges with solar is it is an intermittent resource. I don't have solar at night or when it's very cloudy.

But if I have a stationary battery that I can charge when there's an abundance of solar and then use it when my solar production diminishes, that makes the solar system really powerful.

The combination of solar and batteries becomes very powerful.

Then I can do things like what NEM 3.0 intended, which is, "Hey, don't export any electron in the middle of the day because you'll get paid nothing for it."

Instead, take all the excess electrons you have and store them in your stationary battery or car, or thermally store them in your electric hot water heater.

There are many things. So the battery's role has evolved from providing resiliency to providing what you could call arbitrage. Store free electrons in your battery from your solar system and use them when your solar production diminishes or when the rates are extremely high.

In summer peak rates (between 4 PM and 9 PM), don't buy from the grid because the demand is high, and the rates are extremely high.

Instead, use your own electrons that you stored in the middle of the day. The role of the battery is definitely evolving.

The other role is this notion of participation in helping the grid out. When the signal comes from the network saying, "Hey, help me out, I need a little more supply," your solar may not be there to help you out.

Instead, if I've stored all my excess electrons in the battery, I can discharge my battery and help out the network—and of course, get paid for it.

Molly Wood:
And then I want to ask you about the software controls for this battery management and power management. I don't want to geek out about my energy to that extent. I just want my battery to handle it for me.

Raghu Belur:
Yeah, exactly right. You know, if you step back, what do you want from your energy system? You want to walk up to a light switch, flip it on, and the lights always come on. It's the most economical form of energy. I don't want to pay too much for it. It's the cleanest, and it's the safest. Those are the four things I really care about.

Molly Wood:
Right?

Raghu Belur:
From a software point of view, software is becoming extremely sophisticated.

If you think about it, I have many disparate resources in the home: solar, batteries, EV charging, heat pumps, and the grid. The grid has really complex tariffs now.

I now need sophisticated optimization software looking at all these things economically and deciding what source of electrons to use.

Should I discharge my battery? Should I buy from the grid? Should I sell to the grid? Should I charge my battery or car? Should I discharge my car?

All that optimization has to be done through software. We obviously provide you with all that information on your phone, but you don't have to use it.

Molly Wood:
Mm-hmm.

Raghu Belur:
Sophisticated software should do the right thing for you. If you want to geek out, you're welcome to. If you don't want to geek out on it, that's perfectly okay.

The light switch must always come on, and it should be the cleanest, most economical, and safest form of energy.

Molly Wood:
Amazing. Raghu Belur, thank you so much for the time. This is fascinating. I could do this all day.

Raghu Belur:
Awesome. Thank you, Molly. Great questions. Enjoy being on your show. Bye.

Molly Wood Voice-Over:
So there’s no question we have to get there … the future of energy is distributed - with solar, batteries, EVs, and smart software optimizing it all at the home level WHILE staying connected to the grid.

But HOW we get there is still sort of two steps forward … one step back … the policy and market structures are crucial, too …

My opinion? Prioritize adoption as fast as possible … as cheaply as possible … for as many people as possible ... even if that means subsidizing the crap out of things … I mean we do it for fossil fuels all the time.

More on that topic in the newsletter this week, which you can subscribe to over at everybodyinthepool.com.

That’s it for this episode. Thanks for taking the time to listen. Please rate and review the show if you haven’t already. And if you want to become a subscriber and get an ad-free version of the show, hit the link in the description in your podcast app of choice.

See you next week.