T1 Energy
A Texas factory that assembles solar panels for large solar farms, bought from China's Trina Solar in 2024, and now building its own cell plant.
The scores
Growth1/4 ↑
Sales rose 88% in Q2, but all of 2026's went to Trina, a related party, and gross profit exists only with the credit.
Business quality2/4 →
Module assembly is widely copied; the edge is a credit and a floor that Qcells and First Solar already earn with American cells.
Financial position2/4 ↓
G2 needs $200M-$250M not yet raised, bridged by convertible notes; operations used $103M in the half.
Execution record2/4 ↓
Plant cost up to $510M, first cells pushed to 2027, financing slipped four times, controls weakness unremediated.
Per share capture1/4 ↓
Shares nearly doubled in a year; patents paid in discounted stock, notes convert at $4.46, and new financing may rank ahead.
Potential2/4 →
A large US market and a price floor, but the payoff needs G2 built, financed and the credit kept, two outside T1's control.
Risk profile2/4 →
One external ruling decides the credit that is the profit; Trina is sole customer and main supplier; DOJ and patent suits pending.
Valuation3/4 ↑
About 2.5 to 3.0 times guided 2027 EBITDA prices the plan only partly working, but that EBITDA is the credit.
Evidence quality2/4 →
Figures are filed, but the credit's eligibility, what Trina supplied and the contract prices are unverifiable from outside.
If it works 1.2× · if it fails you keep about 0%.
The Verdict
We rate T1 Energy unattractive. It assembles solar panels in Texas, and all its sales this year went to Trina, the Chinese company that sold it the factory. The business is weak: its profit is a US tax credit, and without it the panels sold at a loss. That credit is denied to companies too close to Chinese firms, and the rules are not yet written. The price is fair if everything goes right, but it is paying for the credit. As a bet it is fragile: if it works, today's money is worth about the same; if it fails, little or nothing is left.
The long caseAmerica keeps building large solar farms, and trade and tax rules reward panels made at home, so a Texas factory can sell what it makes, if its tax credit holds.
The evidence under this - 8 claims
- Management says Q2 margins reflect cost-plus and fixed-margin 2026 contracts written before the 232 floor; module prices after 4 December may be higher (S05).
- The 835 MW is a preliminary figure; the 10-Q does not restate sales volume.
- The same release that set it (March 2026) put first cells in Q4 2026 and G2 financial close early in Q2; both slipped (C17, C18).
- Q2 2026 adjusted EBITDA excluding the tariff refund was about -$13.7M (C05).
- The guide may assume higher module prices after the 4 December floor, so profit before credits need not be negative (C25).
- Not every watt made is sold to a third party within the year, which is when the credit is earned.
- The notice it cites left ownership and effective control, which its restructuring addressed, to future guidance (C11).
- A short seller alleges $65M of Trina cell purchases in Q1 2026 (C15); the 10-Q shows $428.1M bought from the Trina Group in the half without naming the goods (C03).
The company
T1 makes the large glass panels used on solar farms. Its factory near Dallas takes solar cells, the thin silicon wafers that turn light into electricity, wires them into rows and seals them between glass. It can make 5 gigawatts of panels a year.
Buyers are developers building solar farms. They could buy imported panels, or American panels from dozens of other assemblers. What sets panels apart now is less the product than the tax rules around where its parts came from.
The evidence under this - 4 claims
How it makes money
T1 sells panels by the watt, about 30 cents each in the second quarter of 2026, and customers prepay half before each quarter.
On top, the government pays 7 cents for every watt of panel made in America and sold, through a credit T1 sells to financial buyers for cash. That credit, not the panel price, is what turns a loss into a profit.
The evidence under this - 4 claims
The credit is lost entirely if T1 is a specified or foreign-influenced entity, or if the materials in a module fail the material-assistance test (C06, C11).
Where the money comes from, stream by stream
| Stream | Share of sales (H1 2026) | How it earns it |
|---|---|---|
| Modules sold to the Trina Group (related party) | 99.9% | Solar modules assembled at G1_Dallas and sold to Trina, which resells them to US developers as agent for a commission. |
| Modules sold to other buyers | 0.1% | Modules sold directly to US developers; 22.3% of 2025 sales, before 2026 output went to Trina. |
Who runs it
Daniel Barcelo, whose investment vehicle took the company public in 2021, has run it since late 2024. Insiders own about a fifth, almost all one investment fund's stake held by a director; the chief executive owns under 1%. Half his yearly bonus rides on company goals the company does not publish, half on the board's judgement, and his shares simply vest with time, not with results.
The evidence under this - 3 claims
Trina sold 22.5M shares on 21-22 May 2026, after the record date (C16), so its stake is now smaller than the table shows.
Why it wins, and who could beat it
Why customers come back
The advantage is the law. The credit pays American makers per watt, and from 4 December imported panels and cells must pay a minimum price plus a 15% duty.
That is why buyers have signed for panels from a cell plant that does not yet run. But the same floor charges T1 on the cells it still imports, until its own plant starts.
The evidence under this - 4 claims
The threat
Two rivals already have what T1 is building. Qcells makes American cells in Georgia, and First Solar makes panels that need no silicon cells at all. First Solar is also suing T1, saying the cells it imports copy a First Solar patent.
The evidence under this - 3 claims
The rivals: FSLR, JKS, CSIQ, and 1 more that is private, in-house or listed abroad.
Who else does this, rival by rival
| Ticker | Rival | Where it competes | How far along | Priced against it |
|---|---|---|---|---|
| Qcells (Hanwha Solutions), Cartersville, GeorgiaPriced against it: Listed abroad | G2_Austin solar cell fab (Phase 1, 2.1 GW) | On sale | Listed abroad | |
| FSLR | First SolarPriced against it: Yes | G1_Dallas solar module manufacturing | On sale | Yes |
| JKS | JinkoSolar HoldingPriced against it: Yes | - | On sale | Yes |
| CSIQ | Canadian SolarPriced against it: Yes | - | On sale | Yes |
How big it could get
How big this could get
America plans 43.4 gigawatts of large solar farms in 2026, 40% in Texas. At the new import floor of 38 cents a watt, that is about $16.5B of panels a year.
T1 made 935 megawatts in the second quarter, against 9.6 gigawatts installed nationally: about a tenth.
The evidence under this - 4 claims
Who has signed
T1 had 5.4 gigawatts under contract to 2029, most of it with Trina. Two American developers, Treaty Oak and Clearway, signed for 1.5 gigawatts of panels using T1's own future cells. Neither price nor delivery dates are public.
The evidence under this - 3 claims
The results
The numbers
Sales grew 256× in the year to Dec 2025.
Operating margin is −31%.
Running costs take 38¢ of each sales dollar, down from 2727¢ in the year to Dec 2024.
Shareholders' money lost 168% a year, down from −12% in the year to Dec 2023.
It turned 2.2% of sales into free cash, $16.7M over the year.
It owes $208M more than the cash it holds, about 2.2 years of the cash its operations bring in.
The share count rose 24% in 2.0 years, and pay in shares costs 1.5% of sales.
All the numbers
Filed figures from the company's 10-K and 10-Q, read 2026-10-04. Fiscal years end in December.
| Dec 2023 | Dec 2024 | Dec 2025 | |
|---|---|---|---|
| Size and growth | |||
| Sales | $0 | $2.94M | $755M |
| Sales growth | - | - | +256× |
| Profit | |||
| Gross margin | - | 42% | 7.4% |
| Operating margin | - | −27× | −31% |
| Net margin | - | −153× | −49% |
| What growth costs, per sales dollar | |||
| General and admin | - | 26× | - |
| Pay in shares | - | 263% | 1.5% |
| Capital spending | - | 17× | 10% |
| Cash | |||
| Cash from operations | −$87.9M | −$103M | $95.5M |
| Free cash flow | −$276M | −$154M | $16.7M |
| Balance sheet | |||
| Cash and short-term investments | $253M | $72.6M | $182M |
| Debt | $0 | $602M | $390M |
| Net cash | $253M | −$530M | −$208M |
| Returns and shares | |||
| Shares (diluted) | 140M | 141M | 174M |
What growth costs
Overheads have fallen as a share of sales as the Texas plant filled up, but one year is too little to call a trend. The real cost of growing is the cell plant: a large bill not yet financed, paid so far by selling shares and bonds that turn into shares.
The evidence under this - 3 claims
The last report (12 Aug)
Quarter ended June 2026, filed figures.
The second quarter of 2026 looked like progress: sales of $250.1M, up 88%, and a small profit before interest, tax and depreciation.
Look underneath and it was the credit and a $24.4M tariff refund. Without them, the quarter was a loss of about $34M before overheads. Every sale went to Trina, and Trina supplied $428.1M of materials in the half year, more than T1's whole cost of goods.
The same weeks brought news that the cell plant would cost $510M and start in 2027.
What it did to the thesisIt looked like progress but sharpened the central doubt: without the tax credit and a tariff refund the quarter lost money, and Trina was both the only buyer and the main supplier. A costlier, later cell plant added to it, so the view stays Unattractive.
The quarter in detail
Sales growth from the previous quarter, filed figures. The company has restated its own history, so a quarter either side of the restatement may describe a different business; n/c marks one that does not compare.
Earnings per share $-0.14 against $-0.11 expected; $-0.16 as filed, the gap being what the company adds back.
The evidence under this - 5 claims
- Management says Q2 margins reflect cost-plus and fixed-margin 2026 contracts written before the 232 floor; module prices after 4 December may be higher (S05).
- The 835 MW is a preliminary figure; the 10-Q does not restate sales volume.
What management did with the money
Management paid $135M for the cell patents it had licensed, part in shares. But Trina kept a free licence to the same patents. The cell plant has been funded with convertible bonds, loans that turn into shares, while the promised main financing is still unsigned.
Insiders in 2026Q2: 7 purchases and 13 sales, from their Form 4 filings.
The evidence under this - 4 claims
- Schedule E, which lists the assumed licences, was not read; the recital and the definition of Existing Third-Party Licenses are.
- Management values the deal on the licence fees it saves to 2029, which Trina's licence does not change (C14).
The price, and where we disagree
In short. Fair. Cheap if everything goes right, but the promised profit is mostly a government tax credit, and that credit is the thing in doubt.
The shares have fallen 83% from June's high, 41% of it since late September, with no new filing or sales figure, and about a third of the freely traded shares are sold short. All of the earlier fall came from buyers paying less per dollar of sales, not from sales shrinking.
Three dates matter. The import floor starts on 4 December. Treasury still owes the rules on who counts as too close to China. And the 13 November results should show whether T1 has sold its 2026 credits to anyone.
The evidence under this - 5 claims
- 2026-10-27editedre-struck at the 26 October close of $2.12 after a 40.6% fall since the pack
- 2026-10-27editedthe six-month decomposition is dated to the pack; the later fall is stated in C35
How we read the multiple
The whole company, debt included, is valued at about $1.1B. Management says that once the factories run fully, during 2027, it will earn $375M to $450M a year before interest, tax and depreciation. Against that, the price looks cheap.
But at that size the tax credit alone is about $434M a year, more than the middle of that range. So the price is paying for the credit, which shrinks from 2030, ends after 2032, and which the ties to Trina could cost.
Against its own history. The company has no long record of profit to compare with, so there is no normal to measure today's price against.
The evidence under this - 4 claims
- Q3 cash spending, the $73M of Evervolt payments and working capital would lower the cash and raise the figure.
- Restricted cash of $77.3M is not freely available.
- 2026-10-27editedre-struck at the 26 October close of $2.12
- The same release that set it (March 2026) put first cells in Q4 2026 and G2 financial close early in Q2; both slipped (C17, C18).
- Q2 2026 adjusted EBITDA excluding the tariff refund was about -$13.7M (C05).
- The guide may assume higher module prices after the 4 December floor, so profit before credits need not be negative (C25).
- Not every watt made is sold to a third party within the year, which is when the credit is earned.
The 232 floor has no end date and may lift domestic module prices by more than the credit is worth (C25).
Priced against its rivals
The competitors its own annual report names, each confirmed against the rival's filing. Measured on price for each $1 of net worth, cheapest first:
| Ticker | Company | price for each $1 of net worth |
|---|---|---|
| JKS | JinkoSolar Holding | 0.1× |
| CSIQ | Canadian Solar | 0.2× |
| FSLR | First Solar | 1.8× |
| TE | T1 Energy | 2.4× |
The middle of the other 3 is 0.2. T1 Energy Inc is dearest of 4.
Rivals' figures come from a data provider, so read the spread of the group rather than any single pair.
Its rivals are profitable and T1 is not, so the comparison says more about the credit it lacks than about the price.
The bet: what it pays if it works
Think of this as a bet that a Texas solar factory keeps earning a tax credit. Even if T1 sells about three gigawatts of panels a year and the credit holds, the company would be worth roughly what it is today, because the price already pays for that plan.
If it fails, little is left for shareholders: debt and preferred stock are paid first, and without the credit its panels have been selling at a loss.
| EBITDA in two to three years | Gigawatts of modules sold a year needed | Valued like its peers | Like the cheapest |
|---|---|---|---|
| $247.50M (the big win) | about 3 at 100% kept | 6.2× | 1.2× |
| $412.50M | about 5 at 100% kept | 10.3× | 2.0× |
| $412.50M (market forecast) | about 5 at 100% kept | 10.3× | 2.0× |
SEIA and Wood Mackenzie see US installs flat near 44 GWdc a year to 2031; 5 GW, an 11% share, fills G1_Dallas and earns about $412.5M, the plant's ceiling rather than the market's. On that forecast, $412.50M pays 2.0× from today's value: a Fair bet on the market case.
To pay 2× (a Fair bet) it would need about $404.35M of EBITDA (about 5 gigawatts of modules sold a year); to pay 5× (a Cheap bet), about $1.01B of EBITDA (about 12 gigawatts of modules sold a year), at the same multiple.
The steps it has to take
- Done G1_Dallas assembling modules at scale
- Done First third-party offtakes for modules using G2_Austin cells
- In progress 2026 Section 45X credits sold to an unrelated buyer due 2026-11-13
- Slipped G2_Austin's comprehensive financing closes due 2026-11-13
- Slipped G2_Austin ships its first saleable cells into G1 modules due 2027-03
- Not started Treasury proposes the foreign-entity rules and T1 says it stays eligible due 2027-06
- Not started Trina Group below half of net sales in a quarter due 2027-12
- Not started 5 GW of G1 modules running with 2.1 GW of G2 cells due 2027-12
The detail behind the label
The evidence under this - 6 claims
- The same release that set it (March 2026) put first cells in Q4 2026 and G2 financial close early in Q2; both slipped (C17, C18).
- Q2 2026 adjusted EBITDA excluding the tariff refund was about -$13.7M (C05).
- The notice it cites left ownership and effective control, which its restructuring addressed, to future guidance (C11).
- A short seller alleges $65M of Trina cell purchases in Q1 2026 (C15); the 10-Q shows $428.1M bought from the Trina Group in the half without naming the goods (C03).
- Management says Q2 margins reflect cost-plus and fixed-margin 2026 contracts written before the 232 floor; module prices after 4 December may be higher (S05).
- The 835 MW is a preliminary figure; the 10-Q does not restate sales volume.
- The estimate has risen twice in five months (C17).
- Priming would rank new lenders ahead of the $515M of existing convertible notes (C20).
Where we disagree with the market
An American solar champion that new trade rules and the tax credit favour.
Its profit is a tax credit it cannot yet prove it keeps, while Trina is its customer and supplier.
Most analysts rate it a buy, and one broker called the short seller's case against it false. They read adjusted profit with the credit inside it.
The filings show Trina on both sides of the business, as its only customer and its biggest supplier, and nobody outside has yet bought T1's 2026 credits. We would be wrong if an outside buyer buys them and the cell plant is financed with loans rather than shares.
What analysts say, and the evidence
The evidence under this - 6 claims
- Roth: Evervolt is a Singapore conglomerate that bought the IP at auction, and T1 did extensive diligence (S21).
- The short seller profits if the price falls; its invoices are not public.
- The guide may assume higher module prices after the 4 December floor, so profit before credits need not be negative (C25).
- Not every watt made is sold to a third party within the year, which is when the credit is earned.
What could go wrong, and how we would know
The risks
- The credit could be denied. If Treasury's rules treat T1 as too close to Trina, the credit that is its whole profit disappears.
- More shares to come. The cell plant still needs $200M to $250M, and bonds convert into shares at $4.46.
- The credit runs out. It shrinks from 2030 and ends after 2032, when T1 must earn its margin on price alone.
- A patent fight. First Solar is suing T1 over the cells it imports, and Trina keeps a free licence to T1's patents.
- Legal and control problems. Federal prosecutors are looking at a director's 2023 share sales, and the company's own accounting controls are officially weak.
The evidence under this - 8 claims
- The notice it cites left ownership and effective control, which its restructuring addressed, to future guidance (C11).
- A short seller alleges $65M of Trina cell purchases in Q1 2026 (C15); the 10-Q shows $428.1M bought from the Trina Group in the half without naming the goods (C03).
The 232 floor has no end date and may lift domestic module prices by more than the credit is worth (C25).
- Schedule E, which lists the assumed licences, was not read; the recital and the definition of Existing Third-Party Licenses are.
- Management values the deal on the licence fees it saves to 2029, which Trina's licence does not change (C14).
What we still do not know
- What the $428.1M bought from Trina in the first half was, and how many of those were cells.
- What Treaty Oak and Clearway agreed to pay, and when the panels are due.
- Whether Evervolt, the Singapore firm paid $135M for the patents, has ties to Trina, as a short seller claims.
The evidence under this - 3 claims
- Roth: Evervolt is a Singapore conglomerate that bought the IP at auction, and T1 did extensive diligence (S21).
- The short seller profits if the price falls; its invoices are not public.
What would change our view
- Unattractive→NeutralThe credit is proven and the plant financed
It needs all of:
- ○T1 sells its 2026 tax credits to an outside buyer.watched as it happens
- ○The main financing for the cell plant is signed.watched as it happens
- Unattractive→AttractiveThe cells arrive and the customers diversify
It needs all of:
- ○The cell plant ships its first cells into T1's panels.watched as it happens
- ○Trina's share of sales falls below half.checked at the next results, due by 13 Nov
- ○Treasury writes the foreign-ownership rules and T1 says it still qualifies.checked at the next results, due by 13 Nov
What is coming
- 30 October 2026 Evervolt patent instalments of $30M and $18M, cash or stock at a 15% discount: Whether T1 pays in cash or issues more discounted shares, a direct read on the dilution the per-share score rests on.
Every dated event ahead
- 30 October 2026 Evervolt patent instalments of $30M and $18M, cash or stock at a 15% discount: Whether T1 pays in cash or issues more discounted shares, a direct read on the dilution the per-share score rests on.
- 13 November 2026 Q3 2026 results: Whether 2026 credits sold to an unrelated buyer, G2 financing closed, and gross profit excluding tariff refunds turned positive.
- 4 December 2026 Section 232 import price floor on solar cells and modules takes effect: Whether a floor lifts US module prices for T1, or mostly raises the cost of the imported cells it still buys.
- March 2027 First cells from G2_Austin: Whether the cell plant starts on its moved date and budget; another slip strains the financing and the 2027 EBITDA guide.
Since we wrote this
- 27 OctPrice moveShares fell 41% in four weeks with no filing; view heldPrice claims were re-struck at $2.12. The view rests on the credit and the G2 financing, which the 13 November report tests.
- 22 OctNewsPatent payments keep being made in shares; waiver terms unreadThe October payment is in stock at about $3.01. A Commerce waiver was announced but its terms could not be read, so nothing moved.
- 5 OctThesis reviewThe bet now names what it ridesThe owner asked that every bet name the macro trend behind it, after BWXT's bet was missing its chief executive's case for nuclear over gas.
- 4 OctThesis reviewOwner's checklist: revenue streams, cost of growth, management, catalystsThe owner's checklist found the method silent on who runs the company, what growth costs, where the money comes from and what is dated ahead; this pass fills those four blocks from the proxy and filings.
- 29 SepThesis writtenUnattractive, medium convictionThe promised profit is a tax credit whose eligibility is unproven while Trina is both only customer and main supplier, and the cell plant is late, over budget and diluting shareholders.
Also checked 2 times with nothing new to act on.
Sources 35
Sources - 35
Every claim above cites one of these, or states the reasoning that derived it. The quote is frozen at the date it was read: a link can rot, and an agent fetch returns a reading of a page rather than the page. Backs counts the whole claim ledger, so it can name a claim this page does not surface.
- S01primarycompany filing2026-03-31Item 1 business: G1_Dallas and G2_Austin, the Trina Business Combination, the FEOC Restructuring, the IP licence moving to Evervolt, the Module Operational Support and Sales Agency agreements with Trina Solar (U.S.) and their commission terms, the four named competitors, 562 employeesbacksC01C04C12C13C45C30C52C53C57
- S02primarycompany filing2026-08-12one customer at about 100% of H1 2026 sales; sales to and purchases from the Trina Group; deferred revenue from Trina offtakes; 5.4 GW contracted; debt note and conversion prices; preferred stock and Trina anti-dilution right; tariff refunds; DOJ and SEC document requests; RWE suit; First Solar patent suits and the ITC hearing date; material weakness; 45X at about 7 cents a watt; G2 capex of $510M; 294,527,066 shares at 10 August 2026backsC01C02C03C05C07C12C13C14C17C20C21C23C31C32C33C34C36C39C52C53C57
- S03primarycompany filing2026-08-12the statements, adjusted EBITDA reconciliation with the $24.4M tariff refund, 935 MW produced, cash of $156.4M, the 2025 45X credit sale at $0.93, early-stage talks on 2026 credits, Clearway 641 MW, Evervolt IP, G2 first cells in Q1 2027backsC03C05C20C22C24C36C39
- S04primarycompany filing2026-08-123 GW of firm cost-plus and fixed-margin contracts for 2026, unchanged run-rate adjusted EBITDA guidance of $375M-$450M and $650M-$700M, the Evervolt payment schedule, the Hemlock and Corning supply claimsbacksC09C14C23
- S05primarytranscript2026-08-12Q2 2026 earnings call transcript (FMP)data/cache/transcripts/TE-2026Q2.jsonthe analyst questions (Roth, BTIG, Johnson Rice, Bernstein, Needham) on financing timing, Clearway terms, remaining capex, cost pass-through and SG&A; management's answers that remaining Phase 1 capex is $200M-$250M and the financing may 'prime existing debt structures'. The questions are a separate origin from the answers
As it relates to the financing, you guys had talked about end of May and then it was end of June and then end of July. So we're sitting here still kind of mid-August.
backsC09C18C19C24 - S06primarycompany filing2026-07-28about 835 MW of Q2 module sales; adjusted EBITDA of -$14.5M to -$11.5M excluding the tariff refund; G2 capex raised from $425M to $510M; first cells moved from before year-end 2026 to Q1 2027backsC05C17
- S07primarycompany filing2026-07-28$135M total, $2M cash, $60M first tranche in stock, $25M/$30M/$18M due 30 Sep, 15 Oct and 30 Oct in cash or stock at a 15% discount to a five-day VWAPbacksC14
- S08primarycompany filing2026-07-30$120M principal, conversion price about $4.46, 20% premium to $3.72, described as a bridge to the comprehensive financingbacksC18C20C57
- S09primarycompany filing2026-09-29a further $50.0M of the same notes sold to one existing shareholder, closing 30 September 2026, for G2 Phase 1 and general purposesbacksC18C20C57
- S10primarycompany filing2026-03-312.79 GW produced in 2025; G2 remaining capex of about $350M; full financial close targeted early in Q2 2026; first cells in Q4 2026; $160M 45X sale at $0.91; Treaty Oak 900 MW three-year contract; 3 GW contracted for 2026; run-rate EBITDA guidancebacksC09C17C18C24C39C40C41
- S11primarycompany filing2026-08-28the maximum shares issuable on the July notes including make-whole sharesbacksC21
- S12primarygovernment data2025-07-0426 U.S. Code 45X, Advanced manufacturing production credit (Cornell LII) United States Congress4 cents a watt for cells and 7 cents for modules; 75% in 2030, 50% in 2031, 25% in 2032, nil after; no credit for a specified foreign entity or foreign-influenced entity, or a component with material assistance from a prohibited foreign entity
7 cents, multiplied by the capacity of such module (expressed on a per direct current watt basis)
backsC06 - S13primarygovernment data2026-08-11Proclamation adjusting imports of polysilicon and its derivatives, 91 FR 51975, document 2026-16400 (govinfo text) Executive Office of the President, via the Office of the Federal Registerthe minimum import prices, the 15% ad valorem duty from 4 December 2026, and the onshoring program whose tariff offsets the Secretary may rescind
(iii) $0.22 per watt for solar cells; and (iv) $0.38 per watt for solar modules.
backsC25 - S14primarygovernment data2026-02-20the size of the US utility-scale solar build, measured by the government; Texas 40% of it
Developers plan to add 43.4 GW of new utility-scale solar capacity in 2026
backsC27 - S15secondaryindustry data2026-09-14Q2 2026 US utility-scale installations and module capacity near 66 GWdc
The utility-scale segment installed 9.6 GWdc, increasing 61% year-over-year and 56% quarter-over-quarter.
backsC26C28C46 - S16secondaryindustry data2026-06-30US module capacity more than twenty times US cell capacity
there is currently 69.9 gigawatts (GW) of domestic module manufacturing capacity online ... the United States currently has just 3 gigawatts of solar cell manufacturing capacity
backsC26 - S17primarycompany release2026-06-17Qcells, Begins Solar Cell Manufacturing at America's First and Only Vertically Integrated Solar Factory in Cartersville, Georgia Qcells (Hanwha Solutions)a named rival's own account of American cells already in production
by Q3 2026, Cartersville will make 3.3 GW each of ingots, wafers, cells, and 3.5 GW of modules a year.
backsC29 - S18primaryregulator2026-02-12IRS newsroom: Treasury, IRS provide guidance on material assistance from prohibited foreign entities (Notice 2026-15) Internal Revenue Servicethat the February notice is interim and that the definition of a prohibited foreign entity is still to be proposed
Notice 2026-15 PDF provides that Treasury and IRS intend to propose regulations with respect to the definition of a PFE and the calculation of the material assistance cost ratio
backsC11 - S19secondaryanalyst2026-02Mayer Brown, IRS Issues Interim Guidance on 'Material Assistance' from a Prohibited Foreign Entity Mayer Browna law firm's reading that ownership and effective control, the questions T1's restructuring answered, were left open
The Notice largely leaves PFE/Foreign Entity of Concern (FEOC) ownership and broader effective control questions to be addressed in future guidance.
backsC11 - S20secondaryanalyst2026-06-10Fuzzy Panda Research, T1 Energy (TE) - Whistleblower Series: Part 1 - The Solar Cell Invoices Fuzzy Panda Research (a short seller)an interested party's allegation of Trina cell purchases in Q1 2026, and its claim that Evervolt's owner has long ties to Trina
T1 Energy made at least 26 purchases of solar cells from Trina between January 2026 and March 31, 2026, spending at least $65 million.
backsC15 - S21secondarynews2026-05-22the broker rebuttal: Evervolt a Singapore conglomerate that bought the IP at auction, credit accounting standard practice
It is difficult to write a report with '10 Parts' and have every part be misleading and/or false.
backsC15 - S22secondaryvendor data2026-09-29Research pack, frozen beside this thesiscontent/TE/research-pack.jsonprice and range, the move decomposition, FINRA short interest settled 15 September 2026, the analyst split and its drift, the insider window, the cohortbacksC16C35C36C38C41
- S23primarycompany filing2026-05-2622,500,000 shares sold at $8.13 to $9.20, none under a Rule 10b5-1 planbacksC16C55
- S24primarycompany release2026-09-15the ITC complaint T1 was a respondent to is being withdrawn; the district-court suit against T1 continues
its intent to voluntarily withdraw its Section 337 complaint and move to terminate the investigation currently pending before the US International Trade Commission (USITC) without prejudice to refiling at a later date. The Company will continue pursuing its existing TOPCon patent lawsuits in US District Court
backsC31 - S25primarycompany filing2026-07-28the recital that Evervolt granted Trina a royalty-free, fully paid-up, irrevocable, perpetual, worldwide and sub-licensable licence under the acquired IP; T1's assumption of existing third-party licences granted by Evervolt or Trina; the Trina patent suits against Runergy and Canadian Solar that come with the IPbacksC45
- S26primarycompany filing2026-08-0713,615,979 shares issued to Evervolt as the $60M first tranche, registered for resalebacksC14
- S27secondaryindustry data2026-09-14the industry body and its analyst's five-year outlook: US additions flat near 44 GWdc a year to 2031, a 216-240 GWdc safe-harbored pipeline through 2030, the December 2027 placed-in-service cliff and FEOC thresholds stepping up in 2027; read through ir_fetch. Same report and independence group as S15
The safe-harbored pipeline, which will keep tax-credit eligibility, totals up to 240 GWdc, supporting installations through 2030, even after attrition. ... Permitting and interconnection bottlenecks constrain growth to around 44 GW annually, despite rising demand.
backsC46 - S28secondarynews2026-09-21pv magazine USA - U.S. solar capacity nears 300 GW, enough to power 50 million homes pv magazine USASEIA's September 2026 manufacturing count of US module and cell capacity, operating and under construction, as reported by the trade press; read through ir_fetch
Operational U.S. module manufacturing capacity has reached 75.3 GW, with an additional 14.4 GW currently under construction. Domestic solar cell manufacturing infrastructure is also taking shape, with 10.6 GW of operational capacity and 19.1 GW actively being built.
backsC47 - S29primaryindustry data2025-09-09Wood Mackenzie opinion - Outlook for US solar worsens under the OBBBA (Q3 2025 Solar Market Insight highlights) Wood Mackenziethe analyst's own post-OBBBA base and low cases for 2025-2030, and the 48E/45Y deadlines; read through ir_fetch. Co-author of the SEIA report, so not independent of S27
From 2025-2030, our base case outlook puts total solar deployments at 246 GW dc - 4%, or 11 GW dc , lower than our pre-OBBBA outlook. ... Overall, our low case is 18% lower (44 GW dc ) than our base case over the next five years.
backsC48 - S30primarycompany filing2026-07-30the largest US module maker's 2026 volume and adjusted EBITDA guidance with the 45X credits it assumes, its US-made third-quarter volume and its 45.1 GW contracted backlog through 2030backsC49C50
- S31secondaryvendor data2026-10-01FMP key-metrics-ttm and analyst-estimates (annual) for FSLR, CSIQ and TE, fetched through scripts/api/fmp.py Financial Modeling Preppeers' trailing EV/EBITDA and company value over consensus EBITDA for 2027 and 2029, and T1's own consensus EBITDA by yearbacksC50
- S32primarycompany filing2026-05-18beneficial ownership at the 8 May 2026 record date (directors and officers 22.44%, Encompass 19.99%, Trina 17.97%, BlackRock 5.3%); executive officers and tenures; the 2025 annual incentive plan design; the summary compensation table and the January 2025 transaction-bonus RSUsbacksC54C55C56
- S33primarycompany filing2026-10-21how the third $30M Evervolt payment is being met, in shares, and how many are deferred under the 4.99% ownership capbacksC58
- S34secondaryvendor data2026-10-27the 26 October close and the nine-session fall from 3.39 on 9 OctoberbacksC35
- S35primaryregulator2026-10-1573,479,625 shares short at the 15 October settlement against 77,141,996 at 30 September, 4.21 days of average volumebacksC35