Interview with Lux Research Research Associate Matt Feinstein on claims, potentials, cost and perspectives of microinverters and power harvesting optimizers

Matthew Feinstein
Matthew Feinstein

As a Research Associate at Lux Research, Matthew Feinstein currently concentrates in research regarding solar modules, power electronics and other balance of systems technology for Lux Research’s Solar Systems Intelligence practice, as well as system grid interconnection. In the past he has contributed to a broad range of services, including Lux Research’s Smart Grid and Electric Vehicles Intelligence practices. He has also contributed to advisory projects in the Biosciences space.

Mr. Feinstein is the lead author of the Lux Research report "Shorting Out the Myths of Power Electronics, What Fits and What Fails", which was published in February 2011 and examines the claims made by the makers of microinverters and power optimizers.

Solar Server: First off, can you talk about Lux Research's levelized cost of energy model, and how you used this as a tool to evaluate the claims being made?

Matt Feinstein: It is a proprietary model that we developed, along with a major industry player, that I can't disclose. But what it does is that it takes into account system performance including losses, anywhere across the system, for each configuration that we use in the report, as well as other balance of system components and how those might change by configuration. And then it takes into account costs, operating costs as well as any taxes and tax exemptions and things of that sort before calculating a cost per kilowatt-hour for generation.

So It's really a way of measuring the cost of the energy you buy on the system level. Which is also how things like power purchase agreements work. So the same sort of concept, but something that often gets overlooked when you talk about up-front costs.

This takes into account replacement cycle, so if certain inverters claim that they don't need to be replaced over the lifetime of the module, this is where that comes into account.


Solar Server: I'll definitely ask more about these reliability issues later. You state in the report that besides maximum power point tracking, power electronics have few if any technological claims on which they can stand. So in your opinion, how have start-ups based on this technology been able to acquire such significant venture capital, and are these products selling when they get to market?

Matt Feinstein: They are getting venture capital because it is a novel way of looking at distribution architecture for power electronics. Specifically with Enphase, the first microinverter to come on the market, I believe those guys came from Intel, so they certainly had a strong background and were respected in that space. And I think that partially other firms are looking at the fact that they have had some success, success being capturing single digits in terms of market share percent.

That said, that is still around what it is. They are not necessarily skyrocketing, so there are definitely gaining a little bit of ground, but it is modest growth, it is not as though they are taking over the market and string inverters are running to recapture anything.

Before I get a little bit more into that let me explain the few technological claims part. What they boast is that, due to the different distribution architecture they are going to get 15%-25% additional energy harvest. So this is despite the fact that they are going to have a lower weighted efficiency, as in straight DC to AC power, power going in versus power going out. They claim that because of the unique architecture, because of where this inversion is being performed, they are going to get this added energy harvest. But the truth is, that's a claim, that's all that is, often times it's not guaranteed, it's not verified, and frankly many installers don't believe it.

And I've said that more likely the benefits are in the single digits, maybe  under 5% added energy harvest. So that absolutely does not justify the price premium that they are going to incur by installing microinverters, as opposed to a conventional inverter which costs less and to them may or may not be simpler to install or cheaper to install. And maintenance is, of course, another issue.


Solar Server: Viewing microinverters separately, how do these various power optimizers faring?

Matt Feinstein: That's kind of similar, these things are not even doing any inversion. They are claiming that they're going to find the optimal point on a curve. Which is good and often times they are right but it is a matter of how much more energy harvest are you getting with them than without them.

Often times, installers will say we are monitoring the data that we are getting, we are tracking the harvest and it's not nearly what they claim. And it certain situations that really call for it you might see that really good boost of harvest, something like 15%, but for the most part installers are saying, we are seeing less than 5% boost, which doesn't necessarily justify the cost, even for power optimizers, which will be in the ten cent or teens cent range per watt.

The point is that it is not concrete, it is all about the potential. They claim that they won't let faults through, that they will be a lot safer.  Which is fine, but that is also just a question of managing risk on the side of the installer, or whoever is responsible for maintaining the system.


Solar Server: In the report you note that Enphase is definitely gaining market share in residential installations, I think I recall that in recent CSI data, 15% of the solar PV installations of residential scale in California were using microinverters. Do you think that they will continue that growth path, and at some point in the future replace string inverters? Why or why not?

Matt Feinstein: Absolutely not. The reason why is because there is at least one other major inverter manufacturer that is going to come out with a microinverter, and they are certainly going to get a lot more trust from a bankability perspective.

And frankly, there is a break even point, at what system size do microinverters provide benefits or not provide benefits, and installers are fairly confident that they know where that is. And only some are committed to using it and many apparently just want to stay away. I think it is going to be slow growth, modest growth versus a sharp up-tick. Which is something that often comes out in terms of ad-speak.


Solar Server: To go over reliability, you talk a lot in the report about end-component reliability. What concrete steps are manufacturers taking to address that issue?

Matt Feinstein: This is a very wish-washy area, because, since Enphase, a lot of new microinverter companies have come out, swapping out an electrolytic capacitor for a thin-film capacitor, which is a lot more reliable. They are going to take a little hit on efficiency, but they are willing to do so, and they are going to come up for a 25-year warranty instead of the 15-year warranty Enphase offers.

The problem with warranties, particularly inverter warranties, across the board - not just microinverters, is this whole legal nuance. They are tough to judge. And even when you are talking about these companies that are going to offer the 25-year warranties, you are going to pay for it. They are going to have a nice price premium above Enphase, due to both the cost of the capacitor as well as probably the cost of that warranty, the cost of that extra coverage. Plus they are all start-ups, so are they going to be around in 20 years if the thing fails? Who knows.

Sometimes they are trying to find more reliable components to up product reliability. Particularly with the larger inverters, cooling mechanisms are critical. Advanced Energy uses some liquid cooling as well as fans, others only use fans, and microinverters usually use natural convection, which is just air.

I think improving cooling mechanisms is going to be pretty important going into the future to help improve reliability, because it seems like everyone has got the component thing down, plus costs. It's a delicate balance, between how expensive do you want to make these things.

In terms of what other concrete steps they are taking, often times they will show customers their fail rates, things like that, operability, ease of maintenance. They will often have their own LCOE models that they will show which take reliability into account, but these are things that are not public, so it is difficult to comment on them for an industry analyst like myself.


Solar Server: Can you comment further about capacitors and reliability?

Matt Feinstein: So string inverters, central inverters will have some kind of capacitive element that is likely electrolytic. Enphase is currently the only microinverter company that uses an electrolytic capacitor. All the other microinverter firms that I have spoken with use thin film capacitors. But this was never really an issue until we got down to the microinverter level. I think that is when they first realized that they could swap it out.


Solar Server: So what are some of the major trends you see in the next five years regarding power electronics?

Matt Feinstein: I think the hype is going to be dampened. I think we are going to see a lot of the microinverter players fall out of the market. If they are truly viable, incumbents will pick them up, one by one, and some of them will be left out. And many of them will acknowledge that they are only really valuable in smaller installations.

You have a lot of players right now, fighting for market share. I think you are going to see them fall out. I think power optimizers, some might get picked up and packaged with inverters, which is really the way to go. And I think you are going to see a few more of those arise a la Solar Edge and a la Kaco Energy and Tigo. Because that is a good way to reduce costs and shorten the supply chain, and not make installers have to go out and buy two different things.

I think that it is going to continue to be a multiple supplier market, I don't think we are going to see the exclusive agreements, after supplier issues that came about last year and the year before. So I think installers and project developers are going to continue sourcing from multiple suppliers. Which certainly gives people more of a chance to succeed.

People stick with what they know. They are going to stick with the big boys, and the ones that just don't cut it or don't smell right or seem too good to be true, they probably aren't going to make it. Not just on merits of better harvest or anything. You've got to be bankable, you have to stand up on cost and efficiency. And if not, good luck getting your value proposition across.


Solar Server: Anything that we haven't talked about that you think is important for our readers to know?

Matt Feinstein: Particularly with microinverters there is a major value proposition issue that I have, and that is another big part of the reason why I say more modest growth than sharp up-tick. Even if they are going to get you 10% or 15% more energy harvest, as a residential homeowner, do you really care about that? Are you really going to pay twice as much for your power electronics for that? When you could just as easily get a significant amount of your energy bill cut out by having a conventional inverter that the installer might, frankly, like better.

Do you know enough to ask for a microinverter? As long as they don't really care, as long as their incentives are up there, as long as they are mitigating most of their energy bill, while paying as low of an upfront cost as possible. I think for that standpoint going after the residential market is a total wild-card, and probably stands as a significant barrier to its value proposition for microinverter makers.


Conducted by Christian Roselund on February 28th, 2011