From the EditorThere is a pattern running through this issue, and it is worth naming before you read it.
In story after story, the thing that limits the technology turns out not to be the technology. Our cover story finds that electric propulsion broke a thirty-year power ceiling three times in five months — and that not one of those advances was a plasma-physics advance. Each was a power-supply advance. A 20 kW thruster flew because a two-tonne spacecraft unfolded forty metres of solar array to feed it; the constraint moved from the engine to the bus, and the capital is still flowing to the engine. The same shape appears elsewhere in the slate. A transformer shortage turns out to be a shortage of a particular grade of electrical steel. The cheapest new nuclear power turns out to be old nuclear power. And in compute, the collapse in what a token costs is running into a limit that is not price at all but workload — the share of work that cannot move.
None of this is a story about disappointment. It is a story about where to look. When a field's headline number improves quickly, the binding constraint has usually already moved somewhere less photogenic — a factory, a fuel form, a radiator, a supply contract. Twelve features and eight briefings, aligned onto one plane.
— The Ecliptic, Editorial Desk
Cover Story · Frontier & Flight · Engines of Tomorrow
Engines of Tomorrow — The Power Ceiling Breaks: 20 kW in Orbit and the Industrialisation of Electric Propulsion
A 20 kW Hall thruster fired in orbit in March, a 120 kW lithium plasma thruster fired on the ground in February, and a 20 kWe reactor is manifested for Mars in 2028. Every one of those numbers is a power-supply number, not a thruster number.
Electric propulsion spent thirty years stuck below about 5 kW per thruster in flight. In five months of 2026 that ceiling broke three times: K2 Space fired a 20 kW Hall thruster in orbit in late March, JPL fired a lithium-fed magnetoplasmadynamic thruster at 120 kW on the ground on 24 February, and the White House directed NASA in April to field a space reactor of at least 100 kWe for the 2030s. This report asks what actually moved. Building the power, mass and thermal budget from the flown and tested hardware, it finds that none of the three advances is a plasma-physics advance — each is a power-supply advance, and the binding constraint has migrated from the thruster to the bus. K2's 20 kW comes from a 2-tonne spacecraft with a 40-metre solar wingspan, roughly 10 W/kg at the vehicle level; feeding JPL's 120 kW class from sunlight at that intensity implies a vehicle on the order of 12 tonnes near Earth and far more at Mars, which is why a 20 kWe fission reactor is manifested for 2028. The strategic implication is that electric propulsion has quietly become a power-generation and heat-rejection industry wearing a propulsion label — and the capital is flowing to the wrong half of it.
Read the Graphic Edition →
Flagship Beyond Solar
Beyond Solar — The Tandem Haircut: What Survives the Trip From a 35.5% Cell to a Shipped Module
A perovskite-on-silicon cell certified at 35.5% in July. The best tandem module you can actually buy is 24.5%. The 11-point gap is not a laboratory curiosity — it is the entire investment case.
If a perovskite-on-silicon tandem cell is certified at 35.5% and the best silicon module in mass production is 25.0%, how much of that 10.5-point advantage survives the trip to a shipped, warranted product — and what is the surviving fraction actually worth in dollars? This report rebuilds the…
Flagship Joule
Joule — The Transformer Shortage Is a Steel Shortage — August 7, 2026
Grain-oriented electrical steel grew 14.3% last year; power-transformer demand has compounded at 14.0% since 2019. On those two rates the 30% deficit closes in roughly a century.
Why has a four-year transformer lead time persisted through a capital-investment cycle that should have relieved it? This report separates the transformer bottleneck into its assembly layer and its materials layer, and finds that the binding constraint has moved upstream to grain-oriented electrical steel. Global GOES output rose…
Flagship The Gas Gauge
The Marginal Bcf: Power Burn Versus LNG Feedgas Through 2027
The EIA's own July forecast has every cubic foot of 2027 supply growth already claimed — and the claim that cannot flex is the one that shows up for ninety days a summer
Does the United States have enough incremental gas to serve both the LNG buildout and a record power burn through 2027? This report rebuilds the balance directly from the EIA's July 2026 Short-Term Energy Outlook and normalises every claim against the same supply increment. Dry production is forecast…
Flagship The Power Bill
The Power Bill — The Cheapest New Nuclear Is Old Nuclear
Restarting a shuttered reactor costs roughly $2,000–2,600 per kilowatt. Uprating a running one costs more than twice that. Building a new one costs $15,000. The order is not what the policy assumes — and the cheap tier is nearly sold out.
US policy now treats restarts and uprates as one cheap tier of nuclear capacity sitting below new construction. This report prices the two separately from disclosed project costs and finds they are not one tier at all. Restarting a shuttered reactor costs roughly $1,900–2,600 per kilowatt across the…
Flagship Joule
Joule — The Hundred-Hour Battery — August 1, 2026
Industrial heat storage is cheap per kilowatt-hour and expensive per kilowatt — which makes it a siting trade, not a cost trade
Industrial heat storage has an extraordinary cost advantage in the storage medium — the IEA benchmarks thermal storage to 1,000 °C at USD 15–20/kWh, against NREL's USD 334/kWh for a four-hour utility-scale lithium system — and the field has just proven it at scale: Antora and POET commissioned…
Flagship Parameter
Parameter — The Inference Cost Collapse — August 10, 2026
Open weights are 82% cheaper per token, but enterprises report saving only 40–70% — and the gap is the most useful number in the market
Open-weight models now carry a median blended API price 82% below proprietary models — $0.53 against $3.00 per million tokens — yet enterprises migrating to them report realised savings of only 40–60% at moderate volume and 60–70% at scale. This report asks why a linear price advantage produces…
Flagship Parameter
Parameter — The Tokenomics of Inference — July 1, 2026
Why the price of a token is collapsing, the AI bill is exploding, and the margin gap between labs is the story that survives
The unit price of frontier-quality intelligence is falling faster than almost any input in the history of computing — between 9x and 900x per year to hold a fixed benchmark score, with a post-2024 median near 200x per year [1]. Yet enterprise AI bills are rising, not falling;…
Flagship The Reality Gap
The Reality Gap — The Data Centre Went to Space and Sent Back a Bill
Orbital compute is real, flying, and training models. The two most careful cost analyses ever published on it agree the whole idea turns on one number — and it is not a compute number.
TAC-0009 found that beaming solar power from orbit to Earth fails on economics and named one escape route: consume the power in orbit and skip the beam. Nine months on, that route has hardware on it — an NVIDIA H100 trained a language model in orbit, Google radiation-tested…
Flagship Delta-V
Delta-V — LEO Direct-to-Device: Deployment-Rate and Cash-Flow Economics
Six satellites a month, $21-23M each, $3.5B in the bank - the binding constraint is cadence, not physics
Can a low-Earth-orbit direct-to-device operator reach continuous coverage and positive cash flow before its capital runs out? This report builds a transparent techno-economic model of LEO direct-to-cell broadband, anchored on AST SpaceMobile's disclosed numbers and benchmarked against Starlink Direct to Cell and the Amazon-Globalstar combination. The finding: with…
Flagship Chain Reaction
Chain Reaction — The Rent Is $1.8 Million a Megawatt-Year — August 7, 2026
Four of the largest miner-to-AI leases price within 5% of each other on a per-megawatt-year basis. The one investment-grade direct lease prices 32% below them — and that discount is the whole story.
Public bitcoin miners have announced more than $70 billion of AI and HPC contract value, and the sector is being valued on those headline totals. This report converts each disclosed lease into a single comparable metric — contracted revenue per critical megawatt per year — and finds a…
Flagship Chain Reaction
Power Landlords of AI: The Bitcoin-Miner Pivot and the $50 Billion Execution Gap
Public bitcoin miners have signed $110B+ of AI/HPC leases in two years — reclassifying grid access as the asset. Only about a quarter of it has been delivered. This report models the capital-allocation math, the second-order counterparty concentration, and who is exposed.
Can a bitcoin miner become an AI-infrastructure company, and what is that transition actually worth? Public miners have announced more than $110 billion of AI and high- performance-computing (HPC) leases across roughly 17 deals and ~6 GW in two years, with listed miners projected to derive up to…
Joule · 2026-08-10Joule — Briefing — August 10, 2026
1. The US will add a record 86 GW this year, and half of it is solar 2. Lithium's six-month low sits inside a 94% annual gain — the supply response arrived 3. Pack…
Delta-V · 2026-08-10Delta-V — The Maneuver Burden — August 10, 2026
Builds on DLV-0011 — this Briefing takes up that report's open thread on the collision-avoidance operating burden as a hidden opex line and a debris-remediation market catalyst, and tests it against the 2026 environment…
The Reality Gap · 2026-08-09The Reality Gap — The Duration Axis — August 9, 2026
Builds on TAC-0011 — that Briefing argued that AI agents fail on task *duration* rather than task *difficulty*, and left open the construction of a time-to-failure curve across agentic benchmarks. Six months of disclosures…
The Gas Gauge · 2026-08-09The Gas Gauge — Briefing — August 9, 2026
1. **US storage is comfortable and getting more so** — 3,117 Bcf, 195 above the five-year average, and a 3,966 Bcf carry-out in the forecast. 2. **Europe enters the back half of injection season…
Engines of Tomorrow · 2026-08-09Engines of Tomorrow — Briefing — August 9, 2026
1. **Rotating detonation stops being a physics demonstration and starts being a programme** — three separate 2026 test campaigns, one on-orbit contract, and printed hot-section parts on all of them. 2. **CFM's open fan…
Joule · 2026-08-08Joule — Briefing — August 8, 2026
Builds on TIB-0043 — that Briefing separated the post-Hormuz fertilizer shock into a gas-linked nitrogen component and a mined-phosphate component, and left open the question of why the two behaved so differently. The July…
The Power Bill · 2026-08-07The Power Bill — Briefing — August 7, 2026
Builds on CAL-0017 — that report identified the accreditation gap, 86 GW of nameplate additions against declining marginal ELCC, as the real driver of capacity-market scarcity, and asked what it would do to auction…
Star Power · 2026-08-07Star Power — Briefing — August 7, 2026
1. **Pension money arrives** — CFS raises another $1bn, and the buyer class changes. 2. **A record year, and a concentration problem** — $4.48bn raised, 28% of all-time capital in one company. 3. **SPARC…