Joule Heist search

raw · papers · ingested 2026-09-19

Deployment of Enhanced Geothermal System Technology Leads to Rapid Cost Reductions and Performance Improvements (Fervo, Cape Station 2024)

Source: https://eartharxiv.org/repository/view/7665/

Provenance: EarthArXiv preprint, 8 pp., 10 Sept 2024, Fervo Energy authors. Primary field data but vendor-authored and not peer-reviewed; numbers are corroborated by NREL’s 2025 drilling cost-curve update and the ResFrac/URTeC 2025 reservoir paper (both cited below). Retrieved via research agent; full text extracted with pdftotext.

Project Red (Nevada) — the baseline

  • Near-field EGS beside the Blue Mountain hydrothermal plant, north-central Nevada. Hard rock: phyllite, quartzite, diorite, granodiorite. Gradient ~75 °C/km.
  • First horizontal EGS doublet: injector 34A-22 (TD 11,220 ft / 3,420 m), producer 34-22 (TD 11,211 ft), ~8,000 ft TVD, laterals ~3,250 ft (990 m), 9-7/8” hole with 7” cemented casing; vertical observation well 73-22 to 8,009 ft. Spud-to-TD 71 and 58 days for the two horizontals.
  • Multistage plug-and-perforate stimulation with proppant across ~3,000 ft of lateral.
  • 37-day production test (Apr–May 2023): up to 63 L/s, production temperature up to 336 °F (169 °C), max reservoir temperature 376 °F (191 °C), peak 3.5 MWe equivalent. Prior best sustained EGS circulation was ~30 kg/s at Soultz, and that was inflated by flow localizing in a few large faults.
  • Commercial operation from 31 Oct 2023: initial production 341.7 °F, steady ~347 °F (175 °C); 6,200+ hours by Sept 2024 with no thermal decline.

(Primary source for the Red test: Norbeck & Latimer, “Commercial-Scale Demonstration of a First-of-a-Kind Enhanced Geothermal System,” EarthArXiv 5704, 18 Jul 2023 — https://eartharxiv.org/repository/view/5704/. Quote: “Fervo’s horizontal doublet well design is most productive EGS system to-date in terms of flow rate and electric power equivalent.”)

Cape Station (Milford, Beaver County, Utah) — greenfield

  • Adjacent to Utah FORGE. Vertical appraisal well Delano 1-OB to 9,824 ft with permanent DAS + DTS fiber behind casing and a P/T gauge at 8,365 ft; DTS max 444 °F (229 °C).
  • Design basis: target the Granitic Basement at ~8,000–9,000 ft TVD to produce at 390 °F (~199 °C) “to achieve optimal power conversion efficiency with modern organic Rankine cycle (ORC) geothermal power plant technology.” Laterals ~4,700 ft (vs 3,000 ft at Red), drilled at up to 434 °F.
  • Drilling learning curve: Red 71 and 58 days → first Cape horizontal 35 days → recent wells under 20 days. 15 horizontals drilled by Sept 2024 at ~84% of AFE (budget); “100% drilling success rate” on temperature, TD and cost vs plan.
  • Stimulation of the Frisco triplet (one producer between two injectors): 80 plug-and-perf stages across 3 wells, zero screenouts, 95% of stages at design proppant loading; 80% treated at 7,000–9,000 psi, all under the 10,000 psi limit; composite plugs cut drill-out time ~50%; zipper fracs introduced to geothermal.
  • 30-day well test (Jul–Aug 2024): injection WHP 2,000–2,300 psi, production WHP 300–350 psi, self-flowing (no pumps). Initial 120 kg/s settling to ~93 kg/s. Production temperature rose from 369 °F to 382.9 °F (195 °C), no decline.
  • Brine effectiveness 94 kWe per kg/s → 12.0 MWe at 120 kg/s, 9.5 MWe at 93 kg/s; “sustained output of 8–10 MW” from one producer.
  • Heat in place / power density: H = V·ρ·c·(T_r − T_inj); P = η·r·H/Δt with η = plant conversion efficiency and r = thermal recovery factor. With Δt = 30 yr and r = 0.2, the Cape multibench design gives 8.4 MWe/km³ (abstract: 9.1) versus prior DOE estimates of 0.7–0.9 MWe/km³ for 175–300 °C EGS, i.e. 5–10× higher.
  • First 8-well pad targets two formation benches; flexible/dispatchable operation cycles demonstrated (Fig. 12). Cites Ricks et al. 2024 (Nature Energy): with ≥500 MW EGS installed by 2030, learning effects likely make EGS out-compete other firm sources by 2050.

Key quotes

  • “The first production well at the Cape site achieved a peak output of over 12 MW and a sustained output of 8-10 MW.”
  • “we are targeting producing geothermal fluid at 390 °F to achieve optimal power conversion efficiency with modern organic Rankine cycle (ORC) geothermal power plant technology. These design constraints require targeting formation depths of approximately 8,000 ft to 9,000 ft true vertical depth.”
  • “the power capacity density is 9.1 MW per km3, 5-10 times larger than previous estimates for EGS technology.”

Companion: two years of Project Red production data (Fervo blog, 13 Apr 2026)

Source: https://fervoenergy.com/enhanced-geothermal-has-been-proven-at-scale-heres-what-two-years-of-production-data-show/ (vendor blog; only long-duration EGS dataset public).

  • 614+ operating days; uptime 98.4% excluding surface-plant/grid events; zero workovers, remediations or chemical treatments on the downhole system.
  • Average gross 2.1 MW, average net ~1.4 MW at ~36 kg/s and ~347 °F (175 °C); peak 3.5 MWe gross at 60 kg/s.
  • 500+ days with no observable thermal decline, then ~2.5 °F drop; thermal recovery factor ~20% after 600+ days (validates r = 0.2).
  • Stimulated reservoir volume ~975 m lateral × 110 m well spacing × 143 m fracture height. Water recapture ~70% at steady state (near-field site).
  • Cape wells projected ~16 MW gross per well (7× Red).
  • Derived by the research agent, not in source: 2.1 MW / 36 kg/s ≈ 58 kWe per kg/s; against ~420 kWth per kg/s available between 175 °C and a ~75 °C reinjection, gross ORC efficiency ≈ 14%, net ≈ 9%. Cape’s 94 kWe per kg/s at 195 °C implies ~17–19% gross. Consistent with the 10–15% net range in the review literature. Reinjection temperature is not stated in any Fervo document; it enters only through (T_r − T_inj).

Companion: reservoir design (Singh, McClure, Galban, Briggs, Norbeck — URTeC 4245311, June 2025)

Source: https://www.resfrac.com/wp-content/uploads/2025/06/Singh-2025-Fervo-Project-Cape.pdf

  • Why EGS failed for 50 years: bullheading unpropped water into long open-hole sections localized flow into a few fractures; unpropped fracture conductivity is unreliable. Modern fix: cemented casing, plug-and-perf limited entry, tight cluster spacing, large proppant volumes → many closely spaced propped fractures.
  • Benchmarks: Red 63 kg/s (3,000 ft lateral); Cape >100 kg/s (54,000 bbl/day) from three 4,700 ft laterals; Utah FORGE 2024 1,000 ft lateral pair 28 kg/s; historical best ~30 kg/s at Soultz.
  • Base-case model: propped fracture half-length ~670 ft, height ~550 ft; 30 ft cluster spacing clearly better than 45 ft; 15 ft marginal.
  • Well spacing vs thermal breakthrough: 300 ft injector–producer spacing declines rapidly; 600 ft × 200 ft (lateral × vertical) best over 15 yr (~7.2 MW net per producer avg at 900 ft bench spacing); tighter wins over 2–5 yr. Field decision: 500 ft × 200 ft, 700 ft bench spacing, multi-bench “wine-rack” (3 injectors / 2 producers per bench).
  • Net fluid loss 0.36–0.49% of injected volume over 15 years with producers on the pattern’s outside (~200 bbl/day at 50,000 bbl/day).
  • Quote: “Tighter well spacing values result in higher flow rates and greater energy production over the short-term but result in earlier thermal breakthrough and less energy production over the long-term.”

Relevance to the cooling wiki

Fixes the temperature ladder for any downstream heat-driven cooling: EGS brine leaves the wellhead at 175–199 °C, the ORC is designed to take it at ~199 °C, and the reject temperature is unstated but implied to be far below the ≥90 °C a single-effect LiBr chiller needs. Any absorption/adsorption stage would sit upstream of or parallel to the turbine, trading electricity for cold. See the geothermal-driven sorption cooling article.