Bitcoin Heat-Reuse Deployment Registry: Real Projects, Graded by Evidence
Search “bitcoin heat reuse” and you’ll drown in headlines: whole cities heated by mining, 80,000 residents warmed, greenhouses blooming on ASIC exhaust. Some of it is real and some is press-release theatre, and the difference matters if you’re about to spend money. This registry catalogs the actual deployments — pools, district loops, drying kilns, home heaters — and grades each one measured, reported, or PR-only, so you can tell a metered result from a marketing claim. It also names the two headline traps up front, and keeps our own heater line to the same neutral standard as everyone else’s.
Quick answer
Bitcoin heat reuse is real, but it is abundant LOW-GRADE heat, and the honest deployments look nothing like the headlines. Air-cooled ASICs exhaust ~35-45°C air — perfect for warming a room, a greenhouse or drying timber, useless for anything needing steam. Immersion and hydro-cooling reach ~60-75°C water, which is why the strongest real cases are pools and spas (Bathhouse NYC heats four pools and offsets a ~$20-40k/month bill while earning ~1.5 BTC/yr) and district-heat PREHEAT plugged into an existing network (Finland, North Vancouver). Two headlines to distrust: "whole city heated by bitcoin" and "80,000 residents" are the district network's TOTAL population — mining only preheats water that heat pumps finish to distribution temperature. This registry grades every deployment measured / reported / PR-only so you can tell the metered projects from the marketing.
Best-fit, proven: pools/spas, low-temp district PREHEAT in a cold climate, forced-air drying, and single-room residential heating you'd run anyway. Weakest: standalone greenhouse pilots and any consumer heater sold as "pays for itself" — net economics live or die on your electricity price. Canada's cold winters + clean hydro make it unusually well-suited; the marquee examples are still operator-reported, not independently metered.
measured independent published numbers reported operator numbers, plausible PR-only marketing claim, no numbers HPC analog compute heat reuse, not bitcoin 🍁 Canadian
| Deployment | Category | Scale | Heat & economics | Evidence | Status |
|---|---|---|---|---|---|
| Bathhouse (Brooklyn + Manhattan)Bathhouse (spa, self-mines) · New York, USA | water-heating | ~12 ASICs Brooklyn, ~20 Manhattan; Flatiron 14 miners (~2,600 TH/s) heat 4 pools + 3 hammams | Pools held ~40.5°C — a textbook match for immersion-ASIC low-grade heat. Brooklyn power ~$20k/mo, Manhattan ~$40k/mo (roughly unchanged); mining earned ~1.5 BTC in a year (~$90k reported) — heat that would otherwise cost gas/electric is offset while earning BTC | reportedFortune / TIME / MIT Tech Review / DCD | operational2023-2025 |
| Off-the-shelf ASICs have fans removed, are immersed in dielectric fluid, and a heat exchanger pipes the heat into pools (~105°F/40.5°C) and heated marble; miners idle once pools hit target. The best-documented commercial case — $ and BTC figures are operator-stated but consistent across multiple outlets. Strongest 'real economics' example here. | |||||
| MintGreen × Lonsdale Energy Corp 🍁MintGreen + Lonsdale Energy Corp (city utility) · North Vancouver, BC, Canada | district-heating | Deal targets ~100 buildings on LEC's loop (LEC's TOTAL network ~6,500 homes — bitcoin supplies a slice, not the whole grid) | >96% of mining electricity recovered as heat (operator claim); district loop is low-temp hot water, heat-pump boosted. GHG claim ~20,000 t CO2 reduced per MW vs natural gas (operator, not audited); no $ figures | reportedDaily Hive / District Energy / mintgreen.co | operational2021-2026 |
| Immersion 'Digital Boiler' ASIC systems feed recovered heat into LEC's existing hot-water district loop. Announced 2021 as 'world's first city heated by bitcoin'. 🔴 HONESTY: the '6,500 homes' headline is LEC's ENTIRE network capacity, NOT the bitcoin-heat fraction (~100 buildings). No third-party metered delivery numbers. Canada's flagship example. | |||||
| MintGreen × Shelter Point Distillery 🍁MintGreen · Vancouver Island, BC, Canada | industrial | 8× Antminer S9 in a rotating immersion 'tumbler' | Low-grade room/process heat (S9 immersion); no published temperature/kW. unverified | reportedmintgreen.co | operational2020-2026 |
| Eight immersion-cooled S9s heat the whisky barrel-storage room and assist the mash process. MintGreen's first commercial heat-offtake project. A genuine small real distillery deployment (the 'whisky' case). Operator-described only; no measured energy data. | |||||
| MARA Holdings district heatingMARA Holdings (ex-Marathon Digital) · Satakunta, Finland | district-heating | Phase 1 ~11,000 residents; scaled to a claimed ~80,000 across two districts | Air→water at 25-35°C, upgraded to 80°C by heat pumps — mining alone does NOT reach district temp. 'Avoided emissions equivalent to ~700 US homes' since 2024 (operator figure); no $ published | reportedDataCenterDynamics / CryptoSlate / Grist | operational2024-2026 |
| Mining exhaust air passes a heat exchanger warming water to 25-35°C, which district heat pumps boost to 80°C for the municipal grid. Launched June 2024. 🔴 HONESTY: '80,000 residents' is the district network's served population, NOT a metered mining-heat fraction — mining PREHEATS, heat pumps finish. Widely reported but figures originate with MARA. | |||||
| KryptoVault timber (and seaweed) dryingKryptoVault · Hønefoss, Norway | drying | 45 MW facility; heat piped to 12 timber-drying container skips | Warm exhaust air (air-cooled); enough to rapid-dry lumber. unverified (dramatic time-savings claim, no $) | reportedBitcoinist | operational2019-2023 |
| Hydropowered mining exhaust is channeled via corrugated pipes into twelve 'skips' to dry damp timber in ~4 hours vs 2-3 months of sun-drying. Seaweed drying floated as next step. A real, oft-cited industrial drying example. The '4 hours vs 2-3 months' is the operator's — plausible for forced-hot-air kiln drying, not independently measured. | |||||
| HeatCore + MicroBT heat recoveryHeatCore + MicroBT · China + Bellefonte, PA, USA (pilot) | industrial | >300 MW deployed globally by end-2024; HS05 household unit = 5 kW (WhatsMiner M64) | Hydro-cooling (water-cooled) enables higher-grade heat than air; HS05 ~5 kW. unverified | reportedNewswire (MicroBT/HeatCore) | operational2022-2025 |
| Has heated a commercial office and maintained constant-temperature fish-farming pools in China; a US pilot demonstrated fish hatchery, hydroponics and space heating from hydro-cooled miners. Seawater-desalination heating announced for 2025. '>300 MW' is total capacity, not verified heat-delivered. Multiple named use-cases but no independent performance numbers; desalination is announced, not proven. | |||||
| Genesis Digital Assets greenhouse pilotGenesis Digital Assets + RISE / Luleå University · Boden, Sweden | greenhouse | One 600 kW air-cooled container feeding a 300 m² greenhouse | Air-cooled exhaust (warm air, not hot water); 600 kW thermal source. unverified | PR-onlynews.bitcoin.com | pilot2020-2021 |
| A 600 kW mining container's exhaust heat is ducted to a 300 m² greenhouse growing fruit/vegetables; research partners studied fish, insect, algae and fruit-drying extensions. Announced Dec 2020 as an R&D pilot. Heavily press-released in 2020-21 with little follow-up on yields or continuation. Treat as a research pilot of uncertain current status — no harvest/economic numbers ever published. | |||||
| Heata networked server water heatersHeata (spun from British Gas/Centrica) · United KingdomHPC analog | water-heating | Trial: 10 units in British Gas employee homes; broader home rollout | Up to ~4.8 kWh/day into the cylinder (≥2.5 kWh contractual); domestic hot-water temps. Modeled savings up to £340/yr (offsetting electric hot water) or ~£120/yr (offsetting gas); free hot water to host | reportedTom's Hardware / Centrica | pilot2023-2025 |
| A compute server is bolted directly to a home hot-water cylinder; processing heat transfers into the tank for free hot water. Founding prototype was a literal bitcoin-miner-in-oil, since pivoted to general cloud compute. NOW general compute, NOT bitcoin — included because the founding prototype was a bitcoin miner and it's the clearest residential hot-water £-savings data point. Savings are modeled, backed by a real Centrica-confirmed trial. | |||||
| Qarnot 'Digital Boiler' / QBxQarnot Computing · France / Finland / Brescia, ItalyHPC analog | water-heating | ~30 sites across Europe; Brescia phase 1 = 30 QBx units, ~800 MWh thermal/year | Water out at ~65°C; ~95% of compute energy recovered; Brescia ~800 MWh/yr thermal. unverified (ADEME/ENGIE-backed social-housing pilots; no public $) | reportedDataCenterDynamics | operational2010s-2025 |
| Compute modules are cooled by water that exits at ~65°C and feeds building/district hot water. Brescia (with utility A2A) is the flagship district tie-in. HPC/cloud compute, NOT bitcoin — included as the closest proven compute-heat-reuse analog and because its 65°C water shows what liquid cooling unlocks vs air. Its ~800 MWh/yr is one of the few hard thermal numbers in this space. | |||||
| Heatbit Trio / Maxi 🍁Heatbit (consumer product) · Consumer product (incl. Canada) | residential | ~1,500 W/unit; Maxi up to ~39 TH/s, Trio ~10-14 TH/s (only ~400 W drives the chip) | ~1,500 W room heat (~35-45°C exhaust air); heats one room, not a house. Reviews: mining offsets only a small share of running cost; net economics hinge entirely on local electricity price; Trio BTC earnings described as 'negligible' | reportedBob Vila / independent reviews | operational2022-2026 |
| A designer plug-in space heater with an embedded ASIC; nearly all electricity becomes room heat while a fraction of value returns as BTC. Maxi retails ~US$1,249. Independent reviews exist (hence 'reported'). Honest framing: a premium electric heater with a small BTC rebate, not an investment. | |||||
| Canaan Avalon Mini 3 / Avalon Q 🍁Canaan (manufacturer) · Consumer product (incl. Canada) | residential | Mini 3: 37.5 TH/s, 800 W, ~21.3 J/TH; Avalon Q ~90 TH/s | ~2,730 BTU/hr (~800 W); ~80 sq ft primary heat; raises a 10 m³ room 15→21°C in ~10 min. unverified (no independent net-cost study; economics = electricity price minus BTC earned) | reportedHashrate Index | operational2024-2026 |
| A quiet (33-55 dB) home ASIC that doubles as an 800 W space heater — all 800 W becomes ~2,730 BTU/hr. Ships assembled, plug-and-play, with tip-over protection and thermostat cycling. Real shipping product with published specs. Heating figures are manufacturer specs; mining-vs-electricity economics are the buyer's to compute — no neutral third-party ROI data found. | |||||
| D-Central Bitcoin space heaters 🍁D-Central Technologies · Laval, Quebec, Canada | residential | From $235 CAD; S19 edition ~95 TH/s (or ~50-70 undervolted) ≈ up to ~11,090 BTU/hr; BitChimney = single-hashboard S19 convection | S19 edition up to ~11,090 BTU/hr; air-cooled (~35-45°C exhaust); single-room heat. unverified (product spec; net cost depends on local power + BTC price) | reportedd-central.tech (self-reported) | operational2016-2026 |
| Refurbished ASICs (S9, S17, S19, L3+) professionally converted into noise-dampened residential heaters in D-Central's Quebec workshop. The BitChimney is a 3D-printed vertical convection enclosure for silent operation. The registry owner's OWN product line — kept scrupulously neutral. Specs are D-Central's own (self-reported, so not 'measured'). Same honest caveat as every consumer heater: an electric heater with a BTC rebate whose economics hinge on local rates. | |||||
Open data (CC BY 4.0): CSV · JSON · API: /wp-json/dc/v1/heat-reuse-registry
The one constraint that decides everything: temperature
Bitcoin miners produce enormous quantities of low-grade heat, and that single fact explains which reuse cases work and which are fantasy. Air-cooled ASICs — the S9, S19 and virtually every home heater — exhaust air at roughly 35–45 °C. That is genuinely excellent for warming a room, keeping a greenhouse frost-free, or force-drying timber, and it is completely useless for anything that needs steam or high-grade process heat. Immersion and hydro-cooling rigs do better, delivering liquid at about 60–75 °C, which is why the strongest commercial cases are pools and spas (a spa’s ~40 °C water is a textbook match) and district-heating preheat. But even the district projects lean on heat pumps to finish the job: in Finland the mining loop reaches only 25–35 °C, and heat pumps lift it to the 80 °C the grid actually distributes. Bitcoin heat substitutes for a furnace. It does not replace an industrial boiler.
Two headlines to distrust
First: “whole city heated by bitcoin.” The North Vancouver and Finland projects are real and worth celebrating, but the eye-catching numbers — 6,500 homes, 80,000 residents — describe the entire conventional district network’s population, not the slice that bitcoin heat actually covers (in North Vancouver, the deal targets around 100 buildings). Mining preheats water that a much larger heat-pump-and-gas system finishes. Second: any consumer heater sold as “pays for itself.” Independent reviews of the Heatbit and similar units are blunt — the BTC earned is a small rebate, and whether the whole thing is economic depends entirely on your local electricity price. At high rates the mining income is negligible; the honest pitch is “an electric heater that also earns a little Bitcoin,” not an investment. We hold our own Bitcoin space heaters to exactly that standard.
Where Canada fits
Canada is unusually well-suited to heat reuse: long cold winters mean sustained heat demand that would otherwise burn gas or electricity, and provinces like BC and Quebec pair that with cheap low-carbon power. MintGreen’s North Vancouver district loop and Vancouver Island distillery are the leading Canadian examples, and Quebec’s climate underpins D-Central’s own heater workshop in Laval. The honest caveat cuts both ways, though — Quebec’s move to restrict large mining loads on Hydro-Québec works against district-scale in-province heat reuse even as small residential heater use stays unaffected. Model your own numbers with the ASIC heat-reuse calculator, compare the residential options in the Bitcoin heater buyer’s guide, and explore the full landscape at the heat-reuse hub. The pattern across every credible deployment is the same: heat reuse pays when you already needed the heat.
Related products, repair, and setup paths
- how D-Central diagnoses ASIC repairs
- ASIC troubleshooting library
- ASIC manuals and repair guides
- replacement hashboards
- ASIC control boards
- ASIC power supplies
- immersion cooling hub
- home immersion cooling guide
- ASIC miners for immersion planning
- ASIC cooling parts
- airflow shroud before immersion
- compare miner specs in the database
- ASIC repair support
- compare ASIC miner specs
- ASIC miner database
- Antminer S19 specs and profitability
- buy a tested Antminer S19
- Antminer S19 maintenance guide
- Antminer S19 repair service
- Antminer S9 specs
- Bitmain Antminer S9
- Antminer S9 maintenance guide
- S9 hashboard repair parts bundle
- Whatsminer M30S specs
- Whatsminer repair guide
- MicroBT Whatsminer M30S++
- Whatsminer M3x exhaust shroud
- Avalon (Canaan) hashboard repair
- start a repair request
- ASIC repair parts
Last reviewed July 18, 2026.
