Top line numbers. There's there's nearly 80 gigawatts of module production capacity in the U.S., but only about 10.6 of domestic cell manufacturing, with another 19 gigawatts under construction for cells. So this third wave, I think, it's of solar manufacturing is really defined by bringing cell manufacturing on board.

Finlay Colville:
1:17

Who's it helping most? I don't know. Maybe the people that lobbied for it in the first place. I don't believe it's helping the industry as a whole. Right? It's not. You don't create a sustainable manufacturing sector by having a trade policy,

intro:
1:35

the clean energy industry is moving fast. The deals are getting bigger. The technology is evolving, and the stakes have never been higher. Welcome to the Clean Power Hour, the podcast for solar, storage, and microgrid professionals who want to stay ahead of it all? Each week, your host Tim Montague, industry advisor and president of Clean Power Consulting Group, brings you unfiltered conversations with the leaders actually building the energy transition. Now, here's your host Tim Montague.

Tim Montague:
2:10

Today on the Clean Power Hour: Solar Panel Manufacturing in the U.S. of A. My guests today are Ryan Kennedy, senior editor at PB Magazine USA, and Finley Karlville, one of the world's leading analysts of solar manufacturing. We're tackling three big questions: Who's actually manufacturing solar cells and modules in America? How much of that announced factory capacity is real production, and will new tariffs make American solar manufacturing more competitive, or simply make solar more expensive? So you just hosted a conference a week or two ago in Austin, Texas, on the topic of Solar Manufacturing USA, and this was a first event that the two of you put together. What, at a high level, set the table for the listener, for my listener, what is going on in PV manufacturing that solar professionals need to know about?

Ryan Kennedy:
3:08

Yeah, this is this was a really exciting time to put together this this conference. There's a lot of momentum right now in U.S. domestic solar manufacturing. Finley just reported on our website that the solar industry in the U.S. has now cumulatively reached 100 gigawatts of production, and we're now targeting 100 gigawatts of annual production capacity by 2035. This conference was really focused on how do we make these factories work, how do we move from announcements to actual capital expenditure and and bringing equipment online. So Finley, maybe you could talk a little bit about why this year was such a a momentous time to put it together.

Finlay Colville:
3:54

What's different to the situation today is that there are the first investments into solar cell factories, and and that that is the big differentiator because making modules is is a final assembly stage and it's the easiest part of the whole system, and avoid doing all the hard bits, which is making the polysilicon chunks, and then pulling the ingots and slicing the wafers, and then making the solar cells. So, even though the there's there's this current wave of investment at the moment that was stimulated by the Inflation Reduction Act 45x section 45x credits gave the producers a production credit. It was really the module assembly capacity that was what emerged in the past few years. So probably towards the end of last year, I kind of noticed that there was what I call real money flowing into building solar cell factories, and that was probably the turning point where, once you once you start to build solar cell factories, effectively you're in the game in terms of solar cell manufacturing because that's where all the the performance, the technology decisions, the the R and D, that's where it all gets focused, and that's what's different now to the last couple of waves of the U.S. trying to become a domestic manufacturer, it never even got off the ground making solar cells.

Tim Montague:
5:20

I would love it if you guys would paint a picture of where we are today. First Solar is the big dog in the room with, I think, approaching 10 gigawatts of production.

Finlay Colville:
5:33

I don't know. Add another five of that for U.S. specific.

Tim Montague:
5:36

Okay, and and so there's there's thin there's thin film with for solar, and then there's crystalline PV with everyone else. Who are the top credible? Manufacturers of cells and modules in the U.S. today, in your eyes.

Finlay Colville:
5:58

Yeah, I'll kick off on on that one. I mean, I wouldn't like to to judge credibility, but let's maybe talk about who the well. You have a scorecard.

Tim Montague:
6:08

You have a scorecard, Finley. Come on.

Finlay Colville:
6:10

I have a comes out in numbers. There's nothing subjective in there. When the Inflation Reduction Act came in, there was a lot of investment that came in from Asia and particularly China. So a number of the early factories were actually financed for money coming out of Asia and China. You had the likes of you know the the money from Hanwa flowing into QCell's facility in in Georgia, so they're they're one of the major players by virtue of of having multi billions of dollars invested into making not just the modules but the cells that ingots and the wafers, and I would actually put them at the top of the crystalline investment, and you know, and and and production grouping, because they have a very very very large backer in the way of Hanwha Group, and so so they're number one in that. If you want to use the word credibility, because you know it's a very solid financial foundation, and as a company, Q cells has got you know 30 plus years of of making solar cells. If you then look across the whole board, you've got Canadian solar, who are now headquartered in Ontario for a number of years. They had a lot of their manufacturing in China, as we know, and now huge investments into their cell big cell facility in Indiana, and they've got 10 gigawatt module capacity in Texas. So they are one of the biggest now from an investment and capacity stroke production standpoint for solar cells and modules.

Tim Montague:
7:47

Let's round out the picture of who's making what in America. Corning produces polysilicon, ingots, and wafers, and now owns a module factory in Arizona, but doesn't make solar cells. Q Cells is doing the same in Georgia. Illuminate USA assembles modules in Ohio. Texas is becoming a manufacturing hub with Seg Solar, T1 Energy, and Wari. T1 expects to begin cell production in early 2027, and don't overlook Soneva in Georgia and ES Foundry in South Carolina. They're focused on the

critical missing link:
8:20

American-made solar cells. A

Ryan Kennedy:
8:24

lot of my attention was drawn towards those that are making cells right now. Just for some top-line numbers, there's there's nearly 80 gigawatts of module production capacity in the U.S. but only about 10.6 of domestic cell manufacturing, with another 19 gigawatts under construction for cells. So this third wave I think it's of solar manufacturing is really defined by bringing cell manufacturing on board. Yeah, and I think you know a lot of the discussion focuses on different cell technology types. Of are you making Topcon or HJT or you know what's what's your tech? But it seemed to me that the focus a lot of a lot of the leaders were saying it's less important that we're focused on which is going to be the winning technology and more important to focus on building quality production in the U.S. and there's a lot of a lot of equipment sourcing behind that a lot of process learning behind that, and Finley, maybe you could talk a little bit about what you're seeing with the United States learning how to bring on production capacity in a high quality way at the cell stage.

Finlay Colville:
9:49

Yeah, I guess that the U.S. has been so absent from solar cell manufacturing. I mean, 20 years, and 20 years is the time that the solar industry has gone from being a cottage industry to a mass production, you know, 500 gigawatt plus industry. So, if you miss out on that whole learning process, then when it comes to your time to try to become a manufacturing player. Where's the knowledge base? And the knowledge base at the moment is is is absolutely in in Asia. And you know when when Asia and China in particular did all the investments into solar manufacturing, it wasn't just the manufacturing of the solar cells and the modules and the wafers. They basically created an entire ecosystem, so they had all their equipment suppliers, literally at arm's length, working in collectively with them on roadmap, technology, cost reduction, and they also created almost the entire material supply. So right back to the raw materials. So the U.S. is is sitting now trying to. Bring online solar cell capacity, and there's an awful lot of learning that that the country has just not been exposed to. So, and and there are no equipment suppliers in the U.S. that have participated in the Asia boom. I think solar cell manufacturing is a starting point, and it's clear the direction of travel is more and more product manufactured in the U.S. less and less coming from being imported by way of the key the key components, and this did come across a lot at the at the event in Austin. I mean, nobody was you know pushing people that much on on on technology choice. Are they going heterojunction? Are they going top con, and so on? I think genuinely, people wanted to know what's the product, what's the availability of domestic production for solar cells in the next couple of years. You know, can you supply my order book, whatever it's going to be, and is the quality going to be high enough? Is it going? Is the quality going to match the current level of quality and performance that you get from the production lines across Southeast Asia or within China?

Tim Montague:
12:53

So let's talk about the market. There's you you mentioned several times, Ryan, that the U.S. has, you know, 70 to 80 gigawatts of capacity, but that doesn't mean we're making 70 gigawatts, right? The U.S. only installed 45 gigawatts last year. The U.S. is on track to install about the same this year, and so capacity is one thing, but what's actually getting built is a totally other thing is there. Is there is the industry, you know, building factories, both module assembly and cell and module manufacture, in lockstep with what's going on in the U.S. market, or does is that not so important?

Ryan Kennedy:
13:44

Yeah, I think there's definitely sensitivity to market demand, and there was definitely talk of, you know, caution with ramping up actual production and making sure that you don't overbuild and run into, you know, some of the oversupply problems that you might see today with global suppliers, we saw data at the summit showing that you know mature facilities hit 70 to 80% utilization rates, while early stage factories that are still ramping up are hovering between about 15 or 20% utilization. So part of that is, you know, these factories are new and they're learning the learning the processes and building more lines and bringing in more equipment. But I think there's there's some there's sensitivity to where the demand rates are, and you don't want to vastly overbuild capacity and flood the market with more product than it needs. So suppliers are trying to match demand properly. And I,

Tim Montague:
14:56

yeah, go ahead, Finley.

Finlay Colville:
14:58

Yeah, sorry. Yeah, I I kind of look at it a slightly different way. I've always felt that the demand forecasts in the solar industry are woefully on the low side, and typically are. So you actually have to look at what the companies are going to do with that capacity. So company by company, it's always very different. Some of the companies that have got, say, one gigawatt module factories, you know, they may just run them about 100, 150 megawatts in a very kind of like niche rooftop supply, and even go through periods where they're not even having to to turn the production lines on. That's normal. It's not a problem. And there are other companies like First Solar that will sit looking for 97, 98% fleet yield through the entire year because they've got an order going out the next few years that means they can do it, and they know how to run production lines at those utilization rates. So, if we were to add the nameplate capacity of that little one gigawatt player to First Solar, it makes no sense. You're just like apples and oranges. And again, you have to consider if if Elon Musk can justify 100 gigawatts per year over and above what's being forecast by the trade associations, then you have to say that the forecasts are woefully on the low side.

Tim Montague:
16:07

Well, let's talk about Tesla because I hear different numbers, and you just said 100 gigawatts. I thought Tesla was announcing potentially 10 gigawatts of production. What's going on with with Tesla, and what you know, how much truth is there in what Tesla is talking about? Because Elon is famous for making announcements that never come true.

Finlay Colville:
16:32

I'll say a little bit on that one. Okay, so the 100 is just a reference point you did at Davos at the start of the year. That's that's what that one is. You know, it was backed up a second time. So the. Maybe considered to be part of an overall plan within a 10-year period. Then you break it down to, you know, no one, nobody could ever build a 100 gigawatt factory. You know, the drop of a hat, right? Just especially in the U.S. So anything by default is going to be smaller chunks. But even if the first step is 1020, it is for the U.S. market, you know, in another league. If it's 10 gigawatts of ingot wafer cell and module coming online, yeah, that's yep. So it's more than the whole ingot wafer cell that we expect to get built by everyone else within the next, you know, by the end of next year.

Tim Montague:
17:19

So is play tied to data centers in space? Do you think is that a lot of those solar cells are going to make going to space?

Ryan Kennedy:
17:29

Yeah, I mean, I I know that Musk has talked a lot about AI data centers in space being powered by solar, and that appears to be part of the plan. But it seems too early to say, you know, what portion of their panels that they're producing would would be intended for that. They've announced module assembly capacity for the residential space, and they're still making movements there. As as far as the 100 gigawatt goal goes. I think that is a like a headline stated high bar goal, but they they have made real commitments to a factory in Texas. They've been approved by Lamar County for a property tax incentive agreement for a project, 1200 hectare factory in Texas, and the outline of that includes a billion and a half in real property, as well as 8.6 billion in manufacturing equipment and personal property, and they're projecting nearly 10,000 permanent full time jobs, once fully operational, and about 1100 construction jobs. So that's that's in their official tax filing. The timeline for that they're looking at operation 2029 through 2038 is how it was filed in the tax filing, so there's real capital commitment being filed for in Texas, and yeah, it looks like yeah maybe that 10 to 20 gigawatt range just for that one factory there.

Tim Montague:
19:18

Here's a major development since we recorded on October 22nd, Fort Bend County, Texas, approved tax incentives for Tesla's proposed $10 billion solar manufacturing complex near Houston, Texas, codenamed Project Crystal Sun. The facility could manufacture everything from silicon ingots to finished modules, with production targeted for 2029. Tesla hasn't made a final commitment. Meanwhile, SpaceX is pursuing a separate solar manufacturing project near Austin. The ambition is enormous. The question is how much will actually get built. Okay, well let's let's talk about more about the market. There there are a couple of components that we we need to get clarity on, there's the supply chain. You know the the silicon, the ingots, the wafers, which get made into cells, which get made into modules. So there's the material side. There's the machinery that you need to make all this stuff. There's the labor to run these factories, and then there's the electricity to put the lights on inside the factories, which is now a limiting factor for industrial society because of the AI boom and the boom in electrification of everything. It's a both and, but how? What was the temperature? I guess, at the conference regarding this? You know, these these pieces of the puzzle. Does the U.S. have the wherewithal to grow a major solar manufacturing industry?

Ryan Kennedy:
20:57

Yeah, I mean, there was as far as the technical wherewithal. I think Finley could explain that better, but as as far as the read on the floor of the the confidence that I saw the real capital being spent to deploy these factories, the momentum is now. And you know, on the on the trade and policy side of things, there's there's a lot of factors that are feeding this, we have a five-year window remaining now before the key IRA provisions like 45X, 45Y, 45, 48E start to wind down between 2030 and 2032. I was hearing some comments. Confidence on the floor that especially 45x may be extended. You know we we don't have a crystal ball on what'll happen on the policy side, but there seem to be confidence that both sides of the aisle would like to extend manufacturing, domestic manufacturing tax credits. So there's a stable base around that. The big news development before, right before this conference, was Section 232 going into effect. That'll take place on december 4 this year. And what 232 establishes is a 15% tariff on all polysilicon related products coming into the U.S., and then even more importantly, it sets a minimum import price. So there's a price floor for imported products: $21 per kilogram for polysilicon,$100 per kilogram for ingots and wafers, 22 cents per watt for cells and 38 cents per watt for modules. So there's a new price floor for modules coming into the United States, and with that, there's a lot of momentum behind developing them here and supplying them here.

Tim Montague:
23:03

One important distinction on federal tax

credits:
23:05

Section 45(x) supports American manufacturers with credits for solar components beginning to phase down in 2030. The rules for solar projects are different. To preserve eligibility for 45Y or 45E credits, qualifying projects generally needed to begin construction or safe harbor by July 4, 2026, or they must be placed in service by the end of 2027. Those are two very different timelines, and they matter enormously for the industry. Well, and real quick, let's let's let's clarify what that means for for installers who want to buy solar modules or developers. You know, a panel made in Southeast Asia. You might acquire that for 26 cents a watt. Then we have domestic light with you know domestic assembly, but everything else made overseas at maybe 33, and then you get into domestic content with some domestic cells at 44. And if you want to go all in, you're in the 50s or around 50 cents. But the 232 is meant to level that playing field, right, and make domestic content panels competitive with overseas panels, right?

Finlay Colville:
24:28

Yeah, yeah. Effectively, what it's well, it's the several things about 232. To me, fundamentally, is is is is the best way to avoid spending the next 10 years doing anti-dumping CVD cases across the world for every next country that people move, so it's it's it's it's an effective blanket coverage to set a minimum import price. That's that's number one thing. It should allow module makers to have less competition from low-cost modules coming into the country, right? Absolutely puts the pricing up on everything. But what it has done is it's it's put the pricing up on on the solar cells coming in for the module manufacturers in the U.S. where there's not enough solar cell production. So now, if I'm making modules in the U.S., I'm having to buy my solar cells in for 20 something cents and sell them for 40 something cents for the modules. That wasn't the the landscape, you know. Before well, it's the landscape before 232 comes into place. Before then, I could buy my cells in for much lower price, and there was no minimum import price. You sell my modules for a lower price and make a nice margin with my seven cents per watt, 45x on IRA. Who is it helping most? I don't know. Maybe the people that lobbied for it in the first place. I don't believe it's helping the industry as a whole, right? It's not you don't create a sustainable manufacturing sector by having a trade policy. You know, it's it's it's it's it's an open question. But the good thing is that there's mechanisms to change. It's not set in stone. If it looks like it's it's holding back the overall sector from growing, or holding back manufacturing. Then at least it's got that mechanism for change. But it is a little bit alarming to talk about change even before it's coming to effect.

Tim Montague:
26:37

Is the net effect that solar panels are more expensive though for the industry?

Ryan Kennedy:
26:42

Yeah, it's. I saw a figure from Roth Capital that PPA rates are likely going to climb by about $5 per megawatt to absorb the cost increases that that we see from this.

Tim Montague:
27:00

Yeah, so that's my take. Is it's more friction for our industry, but ultimately, solar is unstoppable. I guess maybe we could wrap up this conversation in the next five minutes by talking about that, like some global forces and the need for electricity, the need for cheap electricity, the need for data centers to interconnect their projects as fast as possible, potentially by creating microgrids initially because they don't want to wait three years to get interconnected to the grid. They're doing that with all resources on deck, that includes natural gas, solar, batteries, wind. But what what do you guys see in your crystal ball market wise globally? And you know, I've I've gotten I I have to say I'm walking away from this conversation feeling fairly upbeat that there's a long-term play for U.S. solar manufacturing, and I hope that's you know I hope that comes true.

Finlay Colville:
28:13

Yeah, on on that part, Tim, I I think you're spot on. I mean, there wouldn't be all these discussions and and and people talking about the challenges if this is something that wasn't real, but I think the fact that these people are building factories to make solar cells that are talking about investments coming through into ingots and wafers, this is night and day compared to five years ago in the U.S. But again, it doesn't mean to say it's going to be a success. There's still an awful lot of things that that have to come together on it. On the other one, on the on the the global demand, I don't know, Ryan. I've I've got a few little notes here, but I'll follow in after Ryan. I think.

Ryan Kennedy:
28:55

Sure. Yeah. I mean, things have the the world has changed because of AI data centers, and you know we had relatively flat electricity demand the last 20 or 30 years, a lot of gains in efficiency. You know, zero to 1% annual increase in electricity demand each year. Now that could be 2%, even 4% per year, that total electricity demand is expected to increase over the next couple decades, year over year, and that's compounding and building up. And I think, unfortunately, because of such increases in demand, solar and batteries have been, you know, the energy transition; it it has shifted to more of a complementary role alongside natural gas capacity and other emitting resources. It's not fully a replacement technology right now. The demand funnel is so huge that we are seeing more of that all of the above approach coming in, but with solar and batteries, there's some distinct advantages there. Not just on emissions, but like you mentioned, with microgrids being built, a lot of these data center integrators want to build on private wire and they want to build fast. And solar and batteries are are the fastest way to go. So, on the project side, the the demand curve is just you know shooting right up right now.

Finlay Colville:
30:36

Yeah, the I mean my my kind of take on on on that question, original question, Tim, is that you know when when I look at how solar's been deployed, you know, over the last 2020, 30 years, it has to be said that it's been a it's been a it's been an energy source for people with money. It you know it took off on the back of wealthy homeowners putting solar panels onto their rooftops to to earn money. Then it expanded with institutional money going into solar farms from pension funds and so on, and when we talk about data centers now, you know, a lot. This is Wall Street driven. This is this is this is not, you know, somebody sitting off grid, you know, in the third world who would be dying to get a couple of solar panels outside, you know, to generate their electricity, and it's it's that is such a large untapped part part of the market. And what that needs is a way to unlock financing to make it happen. So so far, it's been a it's it's it's been a financial tool for a lot of people to come in, dip their toe in, and make money. It is. It's unbelievable how quickly you can deploy solar compared to anything else. That's that's probably one of its single bit, single biggest advantages. It can be at the rural level. It can be at the large utility scale level. It can be on rooftops. It can be in ground. Pricing level now is a fraction of what it was in the past, the big issue really for for for for demand taking off is unlocking finance to make it globally accessible.

Tim Montague:
32:12

So, is there anything else from the conference or anything else that keeps you up at night that my listeners should know before we say sayonara?

Ryan Kennedy:
32:22

Yeah, not something that keeps me up at night, but just another thought of something that was shared at the the conference was the the move of perovskites really moving from R and D to active commercialization. There's a lot of excitement around that, a lot of different ways that it's being done, you have integrators like Kalix that are building solar glass, just a glass layer that they put on top of a module. You have Tandem PV with their mechanical stacking method, where they wire their like wire into a junction box, and then you have more like true, true play perovskite integrators like Swift Solar, who just announced at the conference that they had purchased a two gigawatt cell manufacturing of the equipment from Meyer Berger.

Tim Montague:
33:20

Right.

Ryan Kennedy:
33:20

So there's there's real capital being spent on production scale for perovskites, and I think that's exciting. You know, we're getting really close to the theoretical limit of silicon. Long E just announced that they had a cell that achieved 28.1% efficiency, and the theoretical limit is 29.4%. So I think in the U.S. there's there's a great opportunity for higher efficiency, and makes me instead of keeping me up at night, it makes me sleep well, knowing that you know we can push past that theoretical limit of 29% or so, and reach efficiency is closer to 40, about 40% for a two junction perovskite silicon tandem cell. So, on the cutting edge of technical advancement, there's there's a lot of room to gains to be made, and I think that's something to be excited about.

Finlay Colville:
34:21

Yeah, I have to. I sleep in different time zones anyway, Tim. So maybe I'm I'm I'm up at at the time that that no one else has other things to think about. What what does concern me most is this that the market does it does get funds, it does get more factories, but it's just ultimately it's too fragmented to have any coordinated approach, and and then the money just is is is is wasted. It's very fragmented right now in terms of the individual companies in the U.S. and what they're doing. If it becomes more fragmented, then how does that whole thing stack up as a single unit? So that's the thing that that I stand back and look at that that would concern me is that it's not like how how how China was between 2015 and 2324 when it had it it lifted everyone was aligned in what they were doing and building, and the whole supply chain and material supply, and that allowed that entire industry to grow quickly, aligned with the same technology, the same goals, the same cost roadmaps, and it was able to bring the pricing down significantly and increase the performance level way beyond what what anyone else thought was possible, so that's where the U.S. needs to be, and it's it's not there today. So my concern is that it becomes more fragmented, not less fragmented.

Tim Montague:
35:56

And for the U.S. to catch up to a country like China, what is an optimistic point of view: Is it 10 years? Is it 20 years?

Finlay Colville:
36:06

Well, I, I don't, I don't think it's even realistic to say it's going to catch up because there's so many parts of the whole ecosystem that China has has put in place very effectively. You don't press a button and get that. And then the the bigger question beyond that, Tim, is: Does it need to catch up everything, or should it use global knowledge to to move ahead in the way that's more appropriate to the U.S.? Right. So, if it's we're going to catch up, the U.S. is going to catch up on every. Within five years, absolute guarantee it's going to fail. But what points in manufacturing should the U.S. focus on catching up, and how much should it use? Not going to say China knowledge, but external knowledge to benefit the next moves. That's that to me is a bigger question. It's not about you say it's going to catch up by innovation. It's going to be a leader. The U.S. is it's pie in the sky. It's not going to happen. But it's it's it's how you move forward to sustainable solar manufacturing sector. And there are some bits where you might just decide you're buying all these parts and materials from outside the U.S. Right? No, don't waste the time doing that. Do what's needed to get a sustainable industry with the right workforce and factories that are making profits. That's probably, and but I don't think anyone's worked that one out.

Tim Montague:
37:37

All right, I want to thank Finley Karlville with Terra Watt PV Research and Ryan Kennedy, senior editor at PV Magazine, for coming on the show today. Check out all of our content at CleanPowerHour.com. Tell a friend about the show. That's the single best thing you can do to help others find this content. And reach out to me on LinkedIn, gentlemen. Finley, how can our listeners find you?

Finlay Colville:
38:00

Do do the same thing. Just go to LinkedIn, find me. Probably the probably the easiest way.

Ryan Kennedy:
38:08

You can find us on pvmagazineusa.com. We have daily news and analysis. Also coming up at the end of this month, we have our live webinar series. Both of you will be participating in this. On the 20th, we have Finley leading a panel talking about similar topics to what we discussed today, and Tim will be delivering a talk on the 22nd about integrating AI and project workflows. So we're excited to have you both on, and you can learn more about that on PV magazineusa.com.

Tim Montague:
38:44

All right, thank you so much. I'm Tim Montague. Let's grow solar and storage.